The conversation this week is with Scott Litman. Scott is a serial entrepreneur with multiple exits under his belt and someone whom I highly respect and appreciate for what he has done for the technology community here in the Twin Cities. Currently, he is the co-founder, managing partner, and CRO at Lucy.ai. Lucy is an AI-powered knowledge management platform. You see combed through more data in a minute that an entire team could sift through in a year, unlocking, democratizing, and making it instantly accessible to those who need it. Lucy has earned significant recognition and is being adopted by many of the world's biggest ad agencies and Fortune 1000s. Scott also gives back by being on the board of multiple organizations and a co-founder of the Minnesota Cup. The Minnesota Cup is an annual competition that seeks to support and accelerate the development of the best breakthrough ideas from across Minnesota. They are looking for the next generation and the next entrepreneurial success story in our state. More information on the Minnesota cup can be found at mncup.org. Finally, this is a little bit of a blast from the past. But I'd like to tell people that Scott was one of my first bosses after getting out of college. He absolutely had an impact on me at that young age on how to build a technology business through solid culture and leadership.
If you are interested in learning about how AI is being applied across multiple industries, be sure to join us at a future AppliedAI Monthly meetup and help support us so we can make future Emerging Technologies North non-profit events!
Resources and Topics Mentioned in this Episode
Scott Litman 0:00
We're seeing use cases around employee engagement. You know, what do you do when somebody joins a team and they're net new? It's their first time in the company. And there are literally millions and pages of content that existed before they showed up. How are they going to get find out anything? Employee engagement becomes interesting. Similarly, things that are disruptive events like mergers and acquisitions, corporate restructurings, so we start to see Lucy playing a role, where, you know, the company that acquires another company is also acquiring their data they've never ever read and never will read it. Lucy will read all of it. And if they ask that question, they'll find out that they have the answer.
AI Announcer 0:40
Welcome to the conversations on Applied AI podcast where Justin Grammens and the team at emerging technologies North talk with experts in the fields of artificial intelligence and deep learning. In each episode, we cut through the hype and dive into how these technologies are being applied to real world problems today. We hope that you find this episode educational and applicable to your industry and connect with us to learn more about our organization at applied ai.mn. Enjoy.
Justin Grammens 1:11
Welcome everyone to the conversations on Applied AI Podcast. Today we're talking with Scott Litman. Scott is a serial entrepreneur with multiple exits under his belt and someone whom I highly respect and appreciate for what he has done for the technology community here in the Twin Cities. Currently, he is the co founder, managing partner and CRO at lucy.ai. Lucy is an AI powered knowledge management platform. You see combed through more data in a minute that an entire team could sift through in a year, unlocking, democratizing and making it instantly accessible to those who need it. Lucy has earned significant recognitions, and is being adopted by many of the world's biggest ad agencies and fortune 1000s. Scott also gives back by being on the board of multiple organizations and a co founder of the Minnesota cup. The Minnesota Cup is an annual competition that seeks to support and accelerate the development of the best breakthrough ideas from across Minnesota. They are looking for the next generation, and the next entrepreneurial success story in our state. More information on the Minnesota cup can be found at mncup.org. Finally, this is a little bit of a blast from the past. But I'd like to tell people that Scott was one of my first bosses after getting out of college. He absolutely had an impact on me at that young age of how to build a technology business through solid culture and leadership. So for that, I want to thank you, Scott. And of course, thanks for taking the time out of your busy schedule for being on the program today.
Scott Litman 2:25
Absolutely. Justin, thanks for having me. And it is great to see you.
Justin Grammens 2:28
All right, good, good. Well, typically, the way we start these out is, you know, maybe have the guests give a little bit more information about maybe how they got their start and maybe like, what was your passion to get you into this industry?
Scott Litman 2:38
Yeah, so you know, for me, when I was in high school, it was the era where technology was really starting to take off. And the idols of that era were guys like jobs and gates and the things that they had done to found Apple and Microsoft. And so from pretty early age, I was of the mind that I'm going to be a tech entrepreneur, like that's the path. And really, even in a West Coast style, which you know, today, there is no real, there isn't really the same designation. But back then the way that business was done in Silicon Valley was just different. And so that was kind of, you know, the thing I wanted to model after and, you know, I went to the University of Minnesota competed in a national competition. This dates me, but it was designed the computer of the year 2000. And I did this, you know, in, you know, as in the 1987. And the paper I wrote ended up finishing third in the country, I met amazing people through that. A year out of college, I founded my first business, which was an apple and next dealer, and then, you know, three years in, we transition to the internet. And, you know, then the rest was, you know, all kinds of growth and, you know, built the first company sold it built, the second company sold to build the third company sold it built, the fourth company sold it. And all along the way. It was around tech entrepreneurism, building up great teams working generally with Fortune 500 type customers on innovative new technology, you know, kind of the what's next. And our thing in those first four ventures was never to be the Mavericks that were on the bleeding edge that came up with the net new idea that was added to the world, what we're really good at was recognizing an emerging trend that was ultimately being championed by somebody else. And helping businesses adopt that. And the thing that's been so different for us in venture number five with Lucy, is we had to start on more, much closer to that bleeding edge of being the inventors and the creators, versus simply recognizing who are the creators that are going to win, and then helping make those solutions win.
Justin Grammens 4:40
That makes a lot of sense. It's a little bit scary to be on the bleeding edge. I found myself especially with the Internet of Things, most recently of people following me I was really one of the early adopters of IoT and it's, it's still in its infancy, I would say hasn't really, you know, sort of glommed on to what we thought it would be, especially Industry 4.0 and industrial IoT. And, you know, now I'm doing a lot of stuff with AI which I feel is a little bit more mature, but you guys were doing Lucy for many years, right? I mean, I mean, it's 2022 now, but you guys were in there early.
Scott Litman 5:06
So we're six years in. And you know, we're in the third generation of the software and really planning out, you know, the fourth generation, you're always looking forward. And while you know, we're certainly not the pioneers, I mean, you know, there's so much amazing AI work that's been done, you know, by IBM, Google and Microsoft, and a host of others. We were pretty early in on the use cases for the industries we went after. And the things that we were doing, were the first, you know, if a customer was buying us, and it's still true, by the way, even today, if they're buying us it's generally not because they're displacing the prior solution that does it. It's because we're net new, you know, we're going more mainstream, but we still have a bit of an innovators mindset, or a visionary customers mindset that is in play. And certainly in the earlier days, it was exclusively that, you know, now we're, there's enough maturity that we're in that, you know, if you think about the Moore's Crossing the Chasm, you know, we've crossed the chasm, the mainstream is adopting this kind of technology. But that's only been for the last 18 months prior to that it was all very much innovative customers or customers that were willing to be maybe not bleeding edge, but certainly leading edge in their thinking.
Justin Grammens 6:17
Yeah, sure. Sure. Yeah. That innovators dilemma is always very, very difficult to know, how you cross that chasm. You know, you mentioned about industries, maybe could tell me a little bit about you know, I mentioned it during the the intro there. You know, you know, I mentioned ad agencies and stuff like that, is that tell me a little bit about what industry is you're, you're actually going after with Lucy.
Scott Litman 6:35
So when we first started, we went after the agency market, it's a market we know, well, we've got lots of connections and the AI that we've developed, which and Lucy, really the core of Lucy today is AI powered knowledge management, the idea that you can connect to huge volumes of data, largely unstructured data, you know, think about all the PowerPoints, PDF Word documents, and you think about agencies are loaded with them. They're loaded with documents for, you know, marketing strategy and campaign results and Planning, Research mode audiences. And so we started there, and it was a good place for us to start. But ultimately, as we hit the second generation or platform, we realized that, that the primary audience needed to be fortunate 1000, it needed to be large companies with very large teams, and high volumes of content. And the Part of Fortune 1000 We started that was really the research and insights, professional research and insights is an area that many businesses have, they're fairly large teams, they are highly data dependent. And ultimately, they're the subject matter experts that are called on by executive leadership, by marketing, by sales, to better understand trends in the marketplace, to understand their audiences. And to better understand research that's going to drive how products are going to be evolving and sold. We started really focusing there. And that really became a sweet spot. And so we're working with all kinds of customers, whether it's, you know, Pepsi or Warner media GlaxoSmithKline Yeah, target Best Buy at 451, which is data business of Kroger's, and others all around that kind of use case around research and insights, I will say that retail and packaged goods have even larger groups than most in that. And so that really became you know, even the names I just mentioned, you heard a bunch of packaged goods, companies and retailers. That said, as we evolve this business, we're finding more and more use cases where we play a critical role. Because the thing, same thing that is a challenge for the research and insights professional, where they have huge volumes of data. And the same thing that's true in packaged goods and retail is occurring and other use cases and rolls and is incurring in other industries. So we're starting to see Lucy used for sales enablement, because sales teams need to be able to search libraries for proposals, RFP responses, case studies, white papers, customer testimonials, and Company Research. We're seeing use cases around employee engagement. You know, what do you do when somebody joins a team and they're net new, it's their first time in the company. And there are literally millions and pages of content that existed before they showed up? How are they going to get find out anything? And so employee engagement becomes interesting. Similarly, things that are disruptive events like mergers and acquisitions, corporate restructurings, so we start to see Lucy playing a role, where, you know, the company that acquires another company is also acquiring their data data they've never ever read and never will read, but Lucy will read all of it. And if they ask that question, they'll find out that they have the answer. And then, you know, from an industry standpoint, well, retail and packaged goods started, we're seeing the same challenges and healthcare, financial services, pharma, auto, entertainment and other categories as well.
Justin Grammens 9:51
That's amazing. It feels to me like it's very much a complementary technology, right? It feels like you guys are right in this sweet spot where it's like, this is something that a human could never are due. And so it's not like you're just placing workers. Or maybe you are in some cases, I don't know, I just wanted to sort of open that up. And it feels like what you guys are doing is actually sort of getting rid of the monotony and sort of the boring stuff.
Scott Litman 10:12
Yeah, well said, you know, we look at Lucy as being a cognitive companion. It's her job to make life easier for the people that she's assigned to, if you think about it, if I'm working on a proposal, and I need to do research about an audience, and I need to see what have we proposed in the past, and I'm looking for white papers and case studies, the task of looking opening up 10 browsers, going through giant lists of documents, logging into different systems trying to crawl through it and find stuff that's valuable. There's almost zero value for me as an individual to do that. But it's time consuming. But if Lucy can do all of that work, and say, Look, I looked across 40 different systems, I looked across 5 million pages of content. And here are different answers to what you're seeking. Here's page six of the PowerPoint. Here is page 32, of 100 page PDF. Here is a dashboard from Tableau. Here is a research study from a vendor like Mintel. And I look at and go, Okay, now my time is reading those answers, not logging into searching those disparate systems all for the same thing, just to try to get there which could take hours. And so Lucy does in, you know, five or six seconds What would take hours otherwise. And she allows that sales assist person, she allows that market researcher, she allows the product development person, she allows the IT support person to get rid of the minutia that adds no value, and focusing on the task of how do I solve the problem? How do I make the proposal? How do I answer my clients question? How do I do what I'm supposed to do?
Justin Grammens 11:50
And when you mentioned research and insights I was on my mind went right to like gardener, right. So typically, when people would buy gardener and try and find some experts in the field, you guys aren't like trying to try in that angle or whatever, trying to displace them. It feels like the gardener papers would just be loaded into everything you guys have. Right, right.
Scott Litman 12:06
There's a couple of things you've said one, which is, you know, are we displacing people? No, we're complimenting people. The other thing is, we're an overlay on the systems, subscriptions and technology company already has, if a company has a subscription to a third party, like IDC, or E marketer, or you know, insider intelligence are your monitor whoever it is, we don't displace them, we make them better. So with those third parties, we have seen the average usage is 5% of the license users. And why is that? It's because people don't always think to log into those systems and searched them. Or many companies have multiple subscriptions. So the individual has a preference. Oh, I like Gartner. Oh, I like IDC Oh, I like Mintel. And so they start there. And they don't go to System number two or three that the company licenses. Lucy has no bias on that. We connect her to the subscription at IDC, we connect her to the subscription and E marketer that a company has we connect her to the subscription and mental. She searches off. And then the audience, the user says, Oh, I didn't know that Mintel ran something on this, this is exactly what I'm looking for. I wouldn't even thought to go there. So we are an overlay. The same thing is true for internal data, which is an even more primary use case. You know, people will say I can't find anything in teams, I can't find anything in SharePoint. I don't know which SharePoint This is in, is this in SharePoint? Or is this part of the new project that went to office 365, Lucy's connected to all of them. And she searches all of them. Our favorite thing is when somebody says, I didn't know we have this, you know, I didn't know we already did this research. I didn't know somebody already solved this problem. Oh, here are four versions of people have already done what I was about to go do and spend my next two weeks on. I don't need to spend two weeks on this project. I can take the projects that have already been done, and leverage those and maybe I've got a mini project to, you know, synthesize it for today. But that might be a half day exercise versus couple of weeks. It's creating more value from the data a company owns, like what's in SharePoint, or what they license which is going to come from an IDC or Mintel or third party,
Justin Grammens 14:10
Yeah it sounds very powerful. I think I just saw you guys raised your series A maybe week or a couple of weeks ago. Yep. We closed two weeks ago. Yeah. Congratulations. That's awesome. So there, there definitely is a need in in the space. You guys have been sort of self funding this along the way over the past six years.
Scott Litman 14:26
So you know, Dan, and I'ma say, Daniel, my business partner, you know, Dan, but Dan Malin. Dan and I have been connected together, you know, on the four prior ventures, as co founders and co founded the Minnesota cup together. And we're co founders of Lucy together. Well, one of the benefits as a entrepreneur, it's been out there a few times, and it's done is it's easier to raise capital from, you know, your network includes high net worth angels and friends and family that have been past investors and things like that. So we have raised capital, actually, we've raised a fair amount, you know, by Minnesota standards, you know, we're over that 10 million Mark of capital that's been raised. You know, it's been us, it's been our friends and family, it's been high net worth individuals. But the Series A, which was 6,000,001, that we closed the other day was, you know, a professional round it was a venture capitalist was working with Naples ventures, which is out of Naples and Dallas ventures, which is not surprising at a Dallas. And, you know, both groups have been great to work with, they're adding value at the board level, they invested enthusiastically in the business. And it's been great because particularly the closing round happened very, very quickly. And we had multiple suitors that wanted to participate in the round, we actually met with about 10 parties in mid December. And we had multiple that went into due diligence wanting to invest. And it was, our new friends at Naples really sprinted, like through the holiday, they did their due diligence, they took no days off, and they just want to race, but we had other parties that would have gladly invested, that's very different sometimes then, you know, entrepreneurs experience in either first time going through it, it's almost impossible to find, you know, investors, particularly professional ones. And even for entrepreneurs who've done it before, unless a market is perceived as very hot, it can be difficult to raise quickly and efficiently in the challenges. If it takes too long to raise capital, you're spending your precious capital that you do have along the way, you can spend a lot of time as an entrepreneur on the cycles of raising capital that can distract from running the business, so that we're able to close the round quickly and efficiently was a huge benefit to the business as well.
Justin Grammens 16:29
That's awesome. Maybe reflecting back to the mid 90s, I guess right back when you started that first company? Do you think there's something different here around artificial intelligence and what you're doing today and needing to raise money? I guess and needing to spend a lot of money versus I don't know, back then was it just kind of like hanging up your shingle and start a shop somewhere? And and what was it easier? What does it feel like? You know, with regards to the technology like landscape and ways to get into AI today?
Scott Litman 16:57
Yeah, well, there's a couple of really big differences between the eras. One is the landscape for investing is just different today than it was you know, we'll say 20 plus years ago, Minnesota has an ecosystem of investors are numerous entrepreneurs have been successful that go on and then become angels into themselves. So that's a difference. There's a huge difference between trying to raise capital for a services company, and a product company. People don't want to invest in services companies, they don't have the same exit returns as product companies, they are much lower risk. But the reward is much much lower, we've actually been fairly unique in raising the kind of capital we've had for the services businesses in the past, and we've had people actively tell us, I don't invest in services, businesses will invest in you or, you know, we'll, we'll do this thing. But they're just as a very different reward side of it. And oftentimes, services businesses have, you know, they don't risk failing as much, but they risk being non return generating to investors. So, you know, this time, it's a different era. But it's also a different space being product versus services. And it is easier to raise capital for product companies. And then the other thing is just evolution of the leadership team. You know, when I was starting, the only people you could get money from was friends and family, because everybody else is like, what's your track record? You know, today, we have a track record. And that makes it easier. And by the way, it doesn't really create that much insurance for the investor. I mean, we're with Lucy, we have a much greater chance to shoot the moon and do something that goes really big, but we also have a dramatically higher chance and risk of failure than past ventures, either. And, you know, our past track record, de risked things a little bit, but there's still plenty of risk on the products.
Justin Grammens 18:36
Sure, I totally get what you mean about services. I mean, that's where I've been living for many years, actually, most of my companies have been services based ones. And I do have a product platform and AI platform that we're working on me and my co founder called captivation. And it's around presentation skills, essentially using artificial intelligence to track how people are speaking, and are they using positive, uplifting tones, it's looking at their face to make sure that they're looking at the screen, more of an online presentation school tool, I guess, for people to get better, what I'm getting at is there's sort of two sides of the business, right? This is sort of a virtual coach to get you to get you to become better, but we seem to kind of continue to get pulled into the service decide because there's some sort of customization that needs to happen or some sort of hand-holding or, or something that we need to do you guys see in that and Lucy at all, are you getting pulled into some of the services thing? Or can you sort of sell it and walk away?
Scott Litman 19:25
You know, the venture I had prior to this was magnet 360 You know, Salesforce was the maker of software and our business was all around being ideally the best consultancy we could be that would ride on that those coattails and help deploy Salesforce as software. You know, in a perfect world for us we would have we would evolve to the level where people would want to be integrators of Lucy, and we would simply sell a licensed product just like Salesforce sells a licensed product and there'll be somebody else that is getting the services revenue. That would be the ideal state the estate of today is it's really hard to be both a product company and a services company, there's a different mindset, there's a different really financial model to it. And in our earlier days, we tried doing both, even though we knew that it's either hard to do both or even simply a bad idea to do both. But it was just necessity, our product hadn't evolved enough that it could be self service, it hadn't evolved enough that we couldn't meet with a customer and have to listen and listen to their ideas, and then build custom things to make it go. And so on the first, you know, version or to our model, you know, we would tell ourselves, our goal is, you know, 90-10%, license to services. But at the end of the year, we look down and say, Holy crap, we're at 60-40%. You know, that's, that's not what we set out to do. Last year, we were, we were truly 90-10%. And this year will be 9010, or 95. Five, I mean, it is almost entirely license. And the service fees are generally custom integrations and things that we need to do less and less as we create more and more integrations, we have less customization that's needed for that. And that's a really good thing, it really allows you to focus your dev time on the product and making the product better. Now all that said, we have a really phenomenal customer success team. And the customer success team is providing a layer of services that is bundled with the software and you sell it. And they are responsible for working with the customer to make connections to data, they're responsible for working with the customer on how are we going to drive training rollout and user adoption. And so as a product company, I think it is an accepted norm, that customer success is going to be a critical element of delivery and making sure that software is successful. That's an investment we have we have great team and they are baked into the license cost of the software.
Justin Grammens 21:43
Nice. I feel like it can sometimes be hard to turn down that services revenue in the early days. I said, Is that a fair statement? Do you think?
Scott Litman 21:51
That is a very fair statement. Well, I mean, you're looking at it, and you're like, I want to sell anything to anybody. Because I look at the scorecard and like, okay, there's not, you know, I've made commitments to investors and or we're focused on growth, there's got to be dollars coming somewhere. And this sort of fits, and if we just do the services will not only get paid, and that will help us but you know, now we can also get the customer. So yes, it's very, very different in the early stages, and it's almost impossible to say no to that.
Justin Grammens 22:22
For sure. What's a day in the life of a person in your role?
Scott Litman 22:25
For me 80% ID is doing something that is part of my CRO hat, you know, the chief revenue officer. And so I'm either meeting with customers, providing live demos to customers driving strategy with customers, or helping with how are they going to make that transition where they're going to bring in an AI like Lucy, or there's also the participating in events, things like this with podcasts, and the like, you know, the average meeting, or the average day has, you know, 10 to 14 events on the calendar almost entirely through you know, teams, zoom, WebEx, whatever the you know, the technology of the that our audiences, okay, with some days, I feel like the band playing a concert and Cleveland saying good morning, Houston, kind of forget, there's a lot of, you know, doing the same pitch and same messaging over and over again. But I suppose part of being a CRO and a founder is also being the chief evangelist. And you know, getting out in front of audiences as much as possible. The rest of the time is internal working with the team, I think one of the attributes of entrepreneurism and, you know, you're an example of this, you know, as you build a business, you surround yourself with great people, you know, we were all part of matching it together years ago. And that team was awesome. One of the attributes in building a team like that is also putting leaders in place that will run their parts of the business and do it better than if you were in the seat yourself. And so the other 20% of my day is working with a really talented team, whether I'm working with marketing, or customer success, or product Dev, we've got great people, and sometimes providing some mentoring, or coaching or ideation that helps them but ultimately, they are empowered to run their part of the business and make it happen.
Justin Grammens 24:07
Yeah, absolutely. Do you have any thoughts? Or, I guess perspective on where maybe you see AI, going in the next 10 years? You know, you could focus on Lucy, for example, like where do you see Lucy in 10 years or, or any other awesome business applications, you've seen? Think things going, but maybe just think, out into the future? Where do you see this going?
Scott Litman 24:26
I will say that our lens, we look much more at the moment, two and three years, just because we kind of get glued to our own product roadmap. We observe customers part of it as they give feedback, but part of that is you observe how they interact with software. And even if they don't know what they need, you see in their interactions, what they might need, and you pick up on things. Part of it is all of the reading. We do that, you know, you just you're looking for inspiration from what others are doing. Part of it is the inspiration that comes from our own team. We've got really smart people and part of it is visionary stuff that comes from my co founders and I mentioned Dan, but also there's Mark Dispenza, who is our CTO out of New York. So part of it is visionary. And for us, we keep looking at, you know, what is the next two, three year horizon? Part of that is underlying technology? How are we going to make the right bets on the technology components? Where do we make investments that go into the infrastructure, and then part of it is looking at that user experience? For instance, I'll give you an example. One of the things that we observe is that one of the ways that people use Lucy today is to make her a digital subject matter expert within a company. And if you think about, particularly, so much work from home that's occurred over the last two years, you know, watercooler talk has been for the last two years kind of dead. And people go out over slack, or in big companies over teams, and they send out messages and say, Can somebody help me with? Where is this? Or does anybody know the answer to x? Or have we ever done a study on or do we have anything about, and we start have started to see that, in some cases, the number one search engine, inside of a company, is the asking of people, oftentimes through internal social channels, like teams, and slack, we realized, one of our opportunities is people can go to Lucy, and ask that question. But what if we could bring Lucy into that social channel? What if Lucy could just be another person on teams or another member in a Slack community? And so you know, from a roadmap standpoint, that something in you know, over the last year that we've been developing, and we're just about to roll it out, which is really cool, which is awesome, Lucy without adding a new app, to a phone, you know, to an iOS or Android device, is just embedded inside of teams, or slack, which people already use. And since it's hard to get people to download an app, it's even harder to get them to use it. And it's expensive to maintain. All of a sudden, we just have to maintain the connection to teams or slack, they can ask the question within those environments get the answer from Lucy and Olson we are fully mobile and in one of the apps are likely to use that's driven a whole theme of how do we create less friction? How do we create more accessibility of technology? The most important thing with any of this stuff, is that if we're going to build technology is does it get used? We've all been a part of the journey of working with customers that phenomenally use the technology we've deployed for them. But we've also all been a part of customers that bought the technology, and haven't been successful in deployment. It's not just about the technology. It's about how do you reduce barriers to access? How do you get support from people processes and content to make the technology work and succeed? So we constantly look at these things and try to figure out what can we do to make sure that we're more useful? What can we do to make sure that when we inhibit or reduce barriers of access, and what do we do to change ultimately support change of behavior? So there's better usage of of software. So when we think about the future, we think about, it's not just about the technology, it's how are we going to create more access to technology, greater usability? And then the other thing we really look at is, how do we increase the payoff? Like for us, we've done a really good job, I think of developing a solution and Lucy, that is fairly high up on the ability to understand natural language, the ability to understand intents and utterances and apply that against vast quantities of data from different formats and different data types in different systems. And the payoff is either an answer, or answers to the question ranked by Lucy's confidence. And that's kind of the heavy lift that we started with, that we needed to get really good at, then we start to say, Okay, once I have an answer, how can I amaze and delight the user with what I do with it? Now that I have it? How do we make it not just found, but more useful and usable? You know, when we look at the roadmap, it's how do we do a better job of answering the questions, but how do we do a better job of paying off with other features around that, and ultimately, our goal, and if I think about, you know, let's just say that three year out standpoint, is that the technology should be strong enough that Lucy is seamlessly a part of the team, I don't think about going to Lucy, I don't think about going to team, she's just a member of the team. And she's doing things that are indispensable to whatever I'm doing, whether it's I'm an IT support, whether I'm in call center, whether I'm in sales support, whether I'm the new employee that's brought them brought into a team and in Lucy's, my, my favorite friend on the team to start with, because she showed me all the things that didn't know existed, or whether in marketing research and insights, wherever
Justin Grammens 29:39
That is phenomenal. Yeah, I mean, that's where my mind was initially was this virtual employee. And you're right, you gotta get into this mindset of it. It just works, right? This this person just works and they're gonna respond like a human would respond. And then it's just seamlessly happening behind the scenes. And I like the idea of not only providing the answer, but maybe what You're getting into some suggestions, right? in the consumer space, at least, I'm seeing Alexa get better and better. Right? I asked Alexa questions. And then the first was the sort of context and I tell people, I would say, what was the score of the twins game? She knows it's it's the Minnesota Twins, right? She she did that sort of understanding of what I meant, since I live in Minnesota, and I'm a baseball fan, right? But then she's getting better at like, well, Would you also like to know when the next game is right? Giving those sorts of those sorts of suggestions to me as an end user, and I'm just a joke consumer. And a lot of these cases, then makes me want to learn more and ask more, right?
Scott Litman 30:35
That conversational side of AI is really interesting and tricky. And we've been experimenting with a ton. We work with these giant packaged goods, companies. So the other day somebody asked a question on they typed in bread, your packaged goods company, there must be a million pages of content about bread. What do you want to know about bread? In the earlier versions of Lucy, she would just spout out 10 answers about random answers about bread. And somebody would say, Well, that wasn't very helpful. Now, they could help themselves by asking the better question. So then Lucy evolved, we said, well, if people ask something as a keyword, can Lucy follow up? And say, did you find what you're looking for? And if you say, No, tell us what were you looking for? I was looking for Gen Z trends for bread consumption in South Africa. All right. That's a little question. Originally, we would send those queries back to our customer success team, who would then intervene and try to create a moment of amazing delight and, and coach the user and say, Oh, well, I asked this, If Lucy you can do it, too. Now you're gonna get this great answer. But we thought, well, maybe we could skip a step. And instead of sending it to a human, who then looks and sees the quality of the answer, we could take their comment, and actually immediately insert it back into Lucy, and say, I just searched, your feedback. And is this not the right answer? And if so, yeah, we got it. If not, tell us more what you're looking for. We'll help remediate on the back end. But then the next step of that journey is we're not holding a conversation. It simulates a conversation. So we asked a follow up question, but was a pre programmed one. And we had one snap of follow up. So the next thing becomes, we categorize everything, like you made a comment early in the interview to say, it sounds like Lucy does what an army people couldn't do. And one of the things that she does that an army people never could do is she tags, custom metadata, every single page that she reads every frame and clip of every audio and video. So page six of a document is going to have its own tagging of keywords and concepts, and taxonomies. Page seven is going to be different set of that. It allows us to do categorization of data. So when somebody types in bread, Lucy could say these are the best sources that have information about bread, or did you want to know about bread sales? Or bread consumption? Or bread trends? Like what do you want to know? Because I have tagging related to everything on bread, or everything on whatever you're asking about. And so for us, part of the next step is how do we make it more conversational? How do we make it not just a Lucy persona, but actually, she is like that digital assistant that is able to go two or three layers through the conversation, hold the thread of the past interactions and get smarter for it.
Justin Grammens 33:16
Yeah. Oh, man. Sounds like quite a challenge to do. People have been trying to work on this or working on this. For many years AI has gone through this, these AI winters, I guess, where things seem to move forward, but and then they seem to fall back, right?
Scott Litman 33:29
Well, if you think about one of the core values, way back in the management days was constant evolution. And in the magnet 360 business, one of the core values was constant evolution. And in the Lucy business, one of our core values is constant evolution. So even what I just talked through, about going from that recognition of, you know, first it was, somebody asked a question with a poor payoff that was asked a question, and we would take the feedback to customer success, then the next step of evolution was take the feedback and run it through Lucy and see if she can have an answer. And the next step of that constant evolution is going to be how do we make a conversation that pays off where it's Lucy holding the state of that conversation. And so that theme is one that we keep holding to is a core value to kind of drive us forward and keep having us iterate to make things better?
Justin Grammens 34:15
Yeah, that's awesome. That's, that's great. That's a really, really great core value. Because yeah, it lets you constantly evolve, right? And you're never sitting on your laurels, you're always sort of moving in adapting, which is kind of what you got to do in business for anyone looking to enter the field, I guess. Do you have any advice? I guess, like if somebody has a has an idea on how to apply AI, machine learning techniques, I guess. I mean, you guys are well into what you're doing. But yeah, I don't know just having somebody who has built a business here around this space and any suggestions, books, conferences, topics, whatever.
Scott Litman 34:45
I gave very quickly the origin story, you know, to your very first question, but the thing I probably should have added to the origin story is not only did I go to that competition at Apple and compete and which provided some recognition and enhance my network to know P But I also picked up a job when I was in college at Microsoft. Incidentally, Dan had the same job with Mike, same program, same boss, just different college and we didn't know each other at the time. And then, after Microsoft, I helped somebody run a business that was in a very related space to what I would found later on in the first generation of Imaginet. And so for two years, I was general manager of a business. And I think one of the key things is regardless of the quality of idea, if somebody wants that path of entrepreneurism, how do they get entrepreneurial experience? How do they join businesses that are entrepreneurial? How do they learn from those, I get excited at the number of people that were once at imagine that that went on and became founders, you know, whether it's George Reese and Stratos, whether it was Mitch and what he did with code 42, what you've done as a founder, what Bill Fano has done as a founder, what Ron Corp seer, and Tim Solon have done as founders, and Scotty Jesse with smart things, you know, there's a lot of people that drink from the entrepreneurial Well, it's one of my favorite accomplishments is knowing how smart and talented you know, we we picked a lot of great people. And they went out and do Ted brilliant things. And so I you know, the message is if somebody says, hey, I want to be an entrepreneur, how do you get some entrepreneurial experience by joining entrepreneurial businesses just even for a couple of years? Like, what's in the water? How is it different? What are the lessons I learned? The business I was general manager of before I founded imagine it, the founder was really smart guy. But he made so many big mistakes. As an employee, I couldn't help him stop those mistakes. But I learned so much from him. And he also, by the way, he ups that a lot of really smart good things, too. And I learned I learned from both. But it was as an employee being given a paycheck, versus my wiping out my assets. Why I made all those mistakes myself.
Justin Grammens 36:51
Sure, sure. Learn some life lessons on someone else's dime. Ah, yeah. That's, that's great. That's great. No, yeah, I love it. I mean, you just got to sort of dive in and get into the space. And that's kind of what I've done with I have to be frank with artificial intelligence and machine learning is just start the podcast, right? I mean, why wait until tomorrow? Why wait till the perfect time is, you know, just hop on a platform, start reaching out to your network, see, who wants to talk about it and just absorb as much as I possibly can? It's the phase I'm in right now. Well, this is great. Scott, I guess one of the last questions I have here is, you know, how do people reach out to you on LinkedIn? Is that the best place?
Scott Litman 37:28
LinkedIn, certainly even just go to lucy.ai. We have our social channels that are my bios there, things like that. But yeah, LinkedIn is, is probably the easiest place to connect, and at least start things will, you know, I would transition things to email pretty quickly. But LinkedIn is a great starting point.
Justin Grammens 37:41
Awesome. Sounds good. We put liner notes at the end of all this podcast or during the podcast. So all the stuff that we've talked about, and stuff will be in text and transcribed for people to check out and link to and be able to have it be crawled. Maybe even crawled by Lucy, I guess, right?
Scott Litman 37:56
Actually, we take videos like this, and we put them into our file system and connected to Lucy, and they become searched by Lucy. So yeah,
Justin Grammens 38:03
Awesome. Well, there any other topics or things that you know you'd like to share before we wind down here?
Scott Litman 38:08
I'm good. You asked great questions. Fun to catch up.
Justin Grammens 38:12
Absolutely. Scott. Well, no, I appreciate you taking the time. I know you're super busy. You guys are sound like you're in sort of the next the next phase, I guess. And the phases never end. They continue. But you know, the foot is on the gas. We gotta grow. Yeah, well, good. And I'm sure you guys are hiring. Right? If people are looking at the team.
Scott Litman 38:28
So Josh human has joined our team to help us with talent management. And Josh was, you know, of course that imagine that Magna 360. And the say the one thing is the labor market is really challenging right now, it is hard to find people, you look at some of the starting salaries for people coming out of college with no experience. And it's like, wow, we're having to be, you know, really creative on how to build up the team and make sure we continue to find the great people. I think that's going to be a business challenge for almost any business right now making sure that they can find talented staff because there's just such a competitive market out there.
Justin Grammens 38:59
Yeah, for sure. So those that are listening, make sure you Yeah, get into tech, it's a hot area. Specifically, I think this AI machine learning is, you know, it was funny it just to sort of sidetrack here before we ended, but I was interviewing a guy just a couple days ago. And, you know, he just he feels like it's just going to be par for the course, for somebody in software. To understand how AI works, right? It's just going to be, it's going to be ingrained, you're talking about it being built into organizations, I really think is going to be built into them. The ethos of what a software engineer is expected to be, it's not going to be like, Oh, I'm just masters in machine learning, you know, it's really going to have to understand the complexities of how these virtual bots and how AI is going to enter in or sort of intertwine itself with whatever software you're programming.
Scott Litman 39:42
Absolutely. I will say on the labor side, just finding technical project managers, finding QA people, even finding Customer Success reps. They're all in high demand and all hard to find.
Justin Grammens 39:54
Oh, well, we'll be sure to Yep, we'll definitely link off to your guy's career page here with our stuff. So
Scott Litman 39:58
We will take the promotion on that apps. really
Justin Grammens 40:00
Awesome, Scott. Well, thanks again for the time and look forward to talking to you in the future.
AI Announcer 40:06
You've listened to another episode of the conversations on Applied AI podcast. We hope you're eager to learn more about applying artificial intelligence and deep learning within your organization. You can visit us at applied ai.mn To keep up to date on our events and connect with our amazing community. Please don't hesitate to reach out to Justin at applied ai.mn If you are interested in participating in a future episode. Thank you for listening