Episode 156 with Adam Spector

Welcome to Building the Future. I'm your host, Kevin Horik. You can check out new episodes of the show every Tuesday and Thursday at 2 PM. If you missed an episode or want to get more information about the show, please visit buildingthefutureshow.com. Welcome back to the show. Today we have Adam Spector. He's the co-founder and head of business at LiftIgniter. Adam, welcome to the show.

Kevin, thanks so much for having me.

Yeah, man, I'm excited to have you on the show. I think what you're doing is actually really interesting and really cool, and like, it's been something that I think people have talked about if they're in the kind of tech community, but I think you guys are finally bringing this to kind of the average person's kind of company and site and bringing it into kind of just like the average non-technical person's kind of day-to-day kind of online experience. But maybe before we get into that, let's get to know you a little bit better and start off with where you grew up.

Yeah, absolutely. So I grew up in just outside of the Washington, DC area. Which was a fantastic life, great place to grow up, and kind of fun to be around all the things that were happening, positive things that were happening in D.C. at the time.

Sure. So were your parents kind of in politics or not really?

Actually, neither one was. My mom's an artist and my dad's a lawyer. Okay. So neither one of them was. But it's D.C., it's sort of the town of politics, so you kind of, it's around.

Yeah, fair. So you went to university, well, a few universities. What did you take in university and give me the reason behind taking those in university?

I went to Vanderbilt, which I absolutely loved. At Vanderbilt, I had 2 majors, actually. I dual majored in political science, which doesn't lead to a lot of career stuff, but it's a great lead-in to what I did for my graduate degree, which is law, sort of a good law start. And once again, being the D.C. kind of person that I was, politics was interesting. And I like to believe a force for good in many cases. Sure. And then I also did something called HOD, which stands for Human and Organizational Development, which is essentially how to be, you know, business, kind of business administration, leadership, Group dynamics, a lot of different psychology. There are a variety of things that are part of it, but super valuable for really anyone to learn in their career. I certainly loved that course. The last part is I had a minor in philosophy as well.

Oh, interesting.

Lots of things there.

Sure. That's great, man. Then you also went to the University of Miami for business and law?

Yep, I did. Yeah, I got my JD/MBA down in South Florida. Also, you know, fantastic, interesting, totally diverse group of people down there and learned a ton. And then actually took the bar in Maryland, and I'm actually a barred attorney in the great state of Maryland still to this day.

Okay, very cool. So what, what made you go to Miami for current business and law?

A few reasons. First and foremost, I got a little money off for a teammate. That was one. Another one is that my parents and grandparents, and especially my grandparents, live about an hour north near Fort Lauderdale. And at the time they were getting old, and I'm lucky enough to say that my grandfather is still with us. He's actually turning 100 this year.

Wow, that's incredible.

He's even older than he was. Yeah. So yeah, which is super lucky for me, obviously, to still have him around. But I wanted to be close to them. You know, family is a wonderful thing, and especially grandparents, you only get so much time together. So I thought being in South Florida near them would be special, and for 4 years it really was.

No, that's great, man. So walk me through— you're done your education. Walk me through kind of your career history, because you've done quite a lot of things in, in, well, since you've graduated university.

Yeah, it's been a busy time. So yeah, when I first graduated, I took the bar that summer right after graduation, and then I worked at a cool company called Newstar, which was a kind of telecom services provider, uh, working in the corporate development and sort of strategy group there, doing a ton of great stuff in, in kind of just the telecom space in general, which was just interesting and pretty dynamic, um, and just great to learn.

Sure.

Um, in corp dev and strategy was sort of a great combo of a bunch of things I'd done in the past, so it was just exciting, tech-related, which I liked. Um, but then I realized, I actually, another thing I did there which is also I think really important was I helped found there corporate social responsibility group. So, started out what they were doing for how do we, how do we address environmental degradation related to the telecom space.

Sure.

And work towards having a policy around how do we, how do we treat our community, our environment better, while also being really positive and better for business. So, a lot of cool things, enjoyable time there. You know, like a lot of people my generation, and a lot of people today still, I had an itch to do a lot of tech and really get into tech stuff. And, you know, I remember speaking with someone who I got introduced to who worked at a tech startup in the DC area, and he actually told me something that I've never forgotten, which was, he said, if you want to be in politics, be in the DC area. If you want to be in finance, go to New York. If you want to be in entertainment, go to LA. And he's like, and if you want to do tech, you need to go to the Bay Area, San Francisco Bay Area. Sure. And for me, that was sort of a catalyst, and I said, okay, I should start seeing what's out there. And I was lucky enough then to be recruited to come work at a company called Clearwell Systems out in Mountain View, California. They were an awesome eDiscovery company led by some fantastic leaders. Came out there, did eDiscovery with them, part of the business development group, so like a lot of partnerships and sales. Also a great kind of use of my education like JD and MBA, because I understood the legal side, so I understood how lawyers worked, and also kind of the MBA side and business. Short version of that, we got sold for $400 million to Symantec.

That's awesome.

So exciting exit. Yeah, and it was awesome. It was great. It's sort of what you move to Silicon Valley for, is to work on really cool things and maybe make some extra money on top of it.

Very much so, right? Living the dream.

Yeah, so of course, you know, there are always trade-offs, but yeah, I know it's fantastic. Exactly why I wanted to get out here, and it was a great first taste of what Silicon Valley can offer. I mean, how you can really create something from scratch and have it be worth a lot of— a significant amount of money shortly thereafter, because you're creating real value for for your customers, for the community, and everything else. So it's a— it can be a real win, and it really was. So that was that. Then I started my own company. We did identity verification stuff to kind of say, hey, how do we leverage your social networks and value to your own benefit to help you kind of validate who you are, almost like a new form of a FICO credit score? Okay. The company is a cool idea. I still think it's actually one we're seeing a lot of use in today, just I see it around, not what I'm currently doing, but a lot of people are using that idea in different ways, alternative banking and things like that. The company itself, we ended up shutting it down, raised money from investors, returned a pretty good amount of money to investors, not more money, but they got a lot of their original investment back. Right. And we shut it down, and lessons learned there was pretty simple, which was, you know, move fast and find a market. And I guess the one thing I will say is we actually— the one bit of advice I would have from that is if you find a product that you have that people are buying, paying you hard cash for—

Mm-hmm.

To provide that service, Follow the money. That was my big lesson from that. In what we were doing there, we actually, as sort of a corollary to our core technology product, we started doing background checks basically as a service.

Okay.

And had a bunch of customers who signed up to do that. We were making real money doing it, and our core product was not making money. It was sexier from a tech perspective, but we weren't making money doing it.

Got you.

Interesting. Yeah, and so, I mean, the short version of that is we end up, we didn't follow the money. We said we wanted to stick with our core sexy tech product and ignore the money piece, and so the company ended up shutting down. It didn't work out, and then basically a year later, a bunch of other cool companies came out that were kind of doing the same thing, and one of them being Checkr, which is a background check company, which I'm actually an investor in, Okay. And they came out and they completely crushed it, and more power to them. They followed the money, they built a really good service. We easily could have been there far before them, but we didn't do that, so.

Interesting. Yeah, but I think that's really good advice for people listening, right? Like, it sounds kind of probably like now looking back, you're probably like, oh yeah, like, we should have done that. But like, when you're kind of in it too and you're passionate about something, if you're not passionate about it, it's hard. It can be hard, and I get why you made that call, right? But it's a good thing to really consider and tell people, and I love that. So you kind of mentioned a bit of being an investor. How long have you been doing investment, or investing, sorry?

Yeah, so I've had the luck of being, you know, getting introduced to really cool companies, and I've kind of been doing investing, really not a lot of my own money, but Some other people, sort of wealthier individuals who want alternative investments.

Got you.

And don't have the connections in Silicon Valley. Not that I have amazing connections by any means, but I do hear cool deals here and there, and when the opportunity arises, have an opportunity to invest.

Got you. That's really cool. What exactly are you working on now, and why did you decide to co-found LiftLadder? Igniter, sorry.

Yeah, no, not a problem. Um, so yeah, I mean, you know, part of what really got this started, I really have to give credit to my co-founder.

Okay.

Um, Inderneel. How did you meet him?

Sorry, I don't want— I don't mean to interrupt, but I'm always fascinated by how somebody met their co-founder.

Yeah, no, of course, feel free to interrupt. I probably get too excited about things and go a little bit too long. It's good, man. But, uh, but yeah, I mean, so we actually met through, um, the investors in LiftIgniter. So as people were investing in LiftIgniter, um, right at the very early stages, Andrew Neal kind of reached out and said, hey, I really think I need someone on the business side, um, who can help, you know, do a lot of business things, you know, shore up my weaknesses, um, where he kind of knew what his strengths were. Sure. in weaknesses and vice versa, right? So how do I get someone who can help me be great on the business side?

Sure. Well, and even having your legal background is immensely useful.

You're totally right. Thankfully, I feel like I'm finally getting good use of all those years of law school. So yes, I agree.

It's like, you don't— I don't think like we're— like I've worked on at startups and I'm working on a startup now too. It's funny because you do not realize how important having like a good accountant and good legal is until you like have been through it a few times. But like, 'cause you kind of think, well, we're just gonna build this thing, we're gonna launch it. And then like if it takes off and you don't have that stuff in place or have somebody that has any clue, 'cause like I, I have like a design background, so I've never been to law school or accounting and like the classes that I took. Right. that had to do with numbers or any of that stuff. I just was like, boring, never will need this. And now I'm like, oh, I could— should have paid better attention there. So I think that's great that you have kind of a business background and a legal background. I think like that's a huge asset to any startup.

Yeah, no, I thank you for that. And I actually totally agree. I mean, I think having that background is great. I mean, you can build it up, obviously. I mean, you can learn all the same things things that I learned in business school and law school just by doing it, and a lot of people do. But obviously having a little bit more of a formal training is definitely really helpful. It internally saves the company some time and money, because of course lawyers are extremely expensive. So if I can do at least some of that work on our own, my own, it saves us some money and we probably can move a little bit faster and be more nimble as well.

Sure. So what exactly is LiftIgniter?

Yeah, so we are a machine learning personalization company.

Okay. What does that mean for people that don't know what machine learning is? Because I think like, maybe especially non-technical people might not have heard of that, and even probably some technical people.

Totally. Yeah, I know it's definitely a term that is becoming more and more prevalent. And people are starting to hear about it, but don't really understand it yet.

Sure.

And really what that means in our case, right, machine learning personalization is the ability to predict in real time for every single user, but not just user, for every impression. So literally every second and every action you take online, predict your likely next outcome, like the next thing you're going to click on, or next thing you want to watch, or next thing you might want to buy. Sure. And it's actually really to provide a far better, richer, and more valuable experience to the end user. So we like to think of it as a pretty positive thing for everybody. It essentially saves you time finding and discovering what you want. But before we kind of dig into that more, you know, the bigger question is probably one you're asking is around machine learning.

Yeah, totally. So I guess, like, how does that— like, how do you use machine learning in the platform and to provide kind of content for people on the fly because it's— I think to a lot of people it's kind of like this crazy science experiment type thing, but it's actually really, really cool. And I wish people— when I was kind of just earlier today, I rewatched the video on your guys' homepage and I must say that Whoever did that, or like, I'm sure you were involved in it, but like, you guys like nailed it. Like nailed it. How you guys like, like explain that. Like people should just go and like, you know, go, go watch your like quick little video on your homepage. 'Cause like you nailed it. If you're like a startup, like go watch that or, or you wanna learn more about your startup, but like, it's, it's like perfect. Exactly what it should be.

Well, thank you so much. I wrote the whole script for that.

That's awesome.

I can't take credit for the design. We hired an outsourced firm to do that. Sure. But yeah, I wrote the whole script, and I even drew out in my terrible chicken scratch what I theoretically wanted each slide, sort of each segment to look like, and stuff like that. So I'm glad it describes things well.

Yeah, okay. So for people that don't know, what exactly is

it?

Yeah. And I'll stop interrupting you after this.

Oh, no, no, no problem at all. It's good for the flow, probably. But so, you know, machine learning, right, at a high level, it's exactly what it sounds like. It is a machine that is learning, learning on any given application. And what that means is essentially the machine can become dynamic and intelligent and change in real time. So the way I like to describe kind of machine learning, and it really applies at least to the work that we do for our customers, Is imagine, Kevin, the conversation you and I are having right now, right? You're, you're asking questions, I'm responding with answers. And so, and then my answers change your next questions, and so on and so forth, right? That is a fully dynamic, hopefully intelligent conversation that we're having.

Sure.

However, I would completely fail if there were 5 Kevins all on the phone right now asking me different questions. I couldn't scale to do that. I couldn't answer all 5 of you at once.

Sure.

But even if it was just one-on-one, you know, digital properties today give every single user essentially the same static response, right? If you click on, you know, news article A, they will always respond and give you news article B, even if you're someone totally different.

Sure.

Except that's not a dynamic conversation. Right? So what we do is we essentially, we've created this technology that enables like the machine that's running in the backgrounds of our customers' websites to learn from all of their users on that site, and then provide to those, each individual user, a dynamic conversational interaction where it's responding, the machine in the website essentially, or the app, It's responding to every click, every action, everything you do to give you an intelligent response. And then how you interact with that response, just like our conversation, changes the next answer in a sense you get from the website. And the machine will learn from your actions how to tailor the response to you. It improves the model for everybody else also on the site. Just like I've given answers like this before, so I've improved my answers and I've learned from previous discussions.

Sure.

I can give you a better answer today. That's essentially what machine learning does, is it learns from all users. It self-improves itself and can change dynamically as your website or your users are changing.

Yeah. I think that makes sense. Just for people that don't know, you don't really notice anything different on a site. You're just like, wow, this site's so tailored to me. How do they know everything? that I want to read next or view next or see next. It doesn't really matter. That's how I see it. It's pretty cool that— keep going.

Yeah. The thing is, this is the thing that people don't realize is machine learning is already happening. Machine learning and AI, artificial intelligence, kind of termed together a lot of times, and they're similar but different disciplines. But machine learning is happening nonstop all the time, and users don't even know it's happening, just like they don't really know what's happening on our customers' websites either. But, you know, this technology was essentially first developed at Google. My co-founder was actually one of the 5 core people that helped originally build the personalization engine at

Google.

Really? That's cool. It was, and it dramatically changed the amount of time people spent. They dramatically improved the metrics of how much time people spent watching YouTube videos, where they first launched this.

Oh, sure. Of course. Makes sense.

Yeah, you know, because it makes sense, right? If they can do a better, more accurate recommendation of a video, you'll spend more time on YouTube, which is good for the user because they want to find entertaining videos to watch. That's a win for the user. Now, it's also a win, of course, for Google because they make more money if you spend more time on the site. Yeah. So, you know, it's happening on Google, all over the place, machine learning. Facebook, your entire Facebook newsfeed is machine learned.

Totally.

There's, you know, machine learning behind it to decide what videos, what articles, what ads to show you next. And they obviously do a pretty good job making it engaging, because people spend a lot of time on Facebook.

Sure, totally.

So it's really happening everywhere.

Oh, go ahead, keep going.

Yeah, that's all it is. That's it.

Okay. So you guys kind of have— you probably have a better term for it, but kind of like 3 verticals. You'll put content on kind of like a site, and then you do an app side, and then ecommerce. So let's start with kind of just like somebody's website and they just want some content. How do they go about using LiftIgniter?

Yeah, so it's actually pretty straightforward, at least we think. Literally, you add a JavaScript beacon to the site.

Okay.

We start collecting data on the site, and we build our models based on your content. So the articles, videos, or items to buy, let's say, that you have. And then we also learn from your users. And so we gather the data about what your users are doing and how the users understand connections between different pieces of content.

Okay. So do I—

Between those 2 data pieces, we build our models.

So do I have to put my content into your service, or how does that kind of like work to serve up the different things based on who the user is?

Excuse me. So you don't have to do anything extra beyond the JavaScript beacon. We do all the other work, which makes it really easy for our customers to get started. And our system automatically identifies who each user is. And if you remember, we actually don't have any PII, personally identifiable information, about the user.

Okay.

Which makes it very, you know, so we can do this globally because we don't run afoul of privacy rules, or at least we try very hard not to, and hopefully we are not. But because we can use, we use what we would call very sparse data signals.

Okay.

Kind of a machine learning term. Sparse data signals about Each impression, and the real ability and strength of our system is the ability to make a prediction in what we would call real time.

Sure.

So essentially, the moment a user takes an action, we process that action. It's gone through our algorithms, and we've returned back a response to the site. So essentially, we're taking contextual information about what the user is interested in, In this moment in time, and then giving a response, which is super— if you can do that really fast, you can also be extremely accurate and learn from those users right when it's most important and they're most likely to take valuable actions for themselves and for the site.

Sure. No, I think that makes a lot of sense and that's really cool. So how does the app side kind of work?

Yeah. So, app side is actually almost identical.

Okay.

At least on the technology side, right? The data collection, we do a server-side integration.

Okay.

So, you know, there's— you can't collect data with a JavaScript beacon.

Right.

So, we do a server-side data collection.

Got you. Okay. And then, is ecommerce different or how does ecommerce work?

You know, it's funny. We actually don't view ecommerce as particularly different from media or enterprise B2B or anybody else, and here's why. E-commerce has lots of users.

Okay.

They have lots of items, right? Items for sale versus, let's say, an article or a video. And each user needs to be matched up with the perfect product they wanna buy. And that's it. That's literally all it takes, right? They need to be matched up accurately with the right item to buy. So from our algorithm perspective, there is no difference between media or ecommerce. It's just we're a matching engine that's extremely targeted in real time to each impression.

Got you. No, that's actually really interesting. So how does— so I stick this thing on, this JavaScript on my site or one of the other things, and you guys start, you know, sending different content to the different people that come to my site. How do you know, like, what do you start kind of gathering about me? Like, if I've never been to one of your sites that isn't running the JavaScript file, because I'm assuming that if I go to, you know, this site and then that site and they're all running, you know, your JavaScript, you could start learning about me on other people's sites to know what to serve me when I go to another property that's running your service, correct?

So actually, we don't quite work that way.

Okay.

So we very much— we actually silo our data today. So when you go to site A, we're gonna do personalization and recommendations for you on site A, irregardless if we've seen you on site B, C, or D.

Interesting.

Right? We, in a sense, don't track across those right now.

Okay. Is there, like, what's the reasoning behind that? Is that something you're gonna add in the future, or you're just like, no, no, no, we don't wanna do it like that, 'cause it doesn't make sense?

It's possible, but truthfully, we don't need to right now. Okay. The simple answer is we're getting 80 to 100% improvement in click-through rate, engagement, conversion, metrics like that for our customers, without having to merge data across insights.

Interesting, wow, that's incredible.

It is. Thank you. We crush it. We've never lost an A/B test against anybody.

Yeah. Wow. Pretty cool tech. Yeah. Okay. So are you pulling data from any other kind of like my social stuff or any other kind of sites like shopping or anywhere else, or is it still just kind of very much just your vertical? Or my site, sorry.

It's very much just on that site. Okay. We're not pulling data from any third-party sites or anything else. It's literally What are you doing on that site at that moment in time?

Okay, interesting.

Yeah, people wouldn't recognize that as being really powerful, but it actually is extremely powerful. Once again, as long as you're using the data in the right ways.

Right, okay.

The fact that you liked Coca-Cola on Facebook doesn't actually have a lot of predictive power for the fact that you're going to want to buy that t-shirt on this other t-shirt site or something.

Yeah, okay, that's interesting. No, that makes some sense. But like, you could potentially sell me other Coca-Cola down the road or like tomorrow, or not really?

Well, you could, but here's the thing, right? It just, you know, yeah, we know you maybe have an affinity to Coca-Cola, but that doesn't actually mean you want to buy Coca-Cola today. What's far more important is, you know, what are you looking to buy right now, right? You're in Canada, it's cold. You're probably looking to buy, you know, a winter coat, let's say, today. Sure. However, maybe you then go on a website, a travel site, let's say, and you're looking for vacations to the Caribbean, right? Because you want to go somewhere warm. You know, now instead of looking at what you used to like previously or your past purchase history, we'd be far smarter if we said, hey, hey, Kevin, you should go look for this pair of shorts or this great new bathing suit. Because, you know, right now you're really interested in that warm weather.

Got you.

Who cares what you did previously, that you looked to buy a coat earlier today? Now you're looking for something warm, so let's respond in real time to who you are at that moment.

Yeah, okay, that makes total sense, right? Because you're right, if I'm looking for a vacation, especially if it's in the future, you don't care about the past. Yeah, that's interesting. So kind of for people that don't understand, how do you guys even kind of do that?

Yeah, so it's a super detailed ensemble of a variety of different machine learning algorithms.

Okay.

Almost all of which we've homegrown here, but of course, we're built like, frankly, everything else in the world. We've grown on the backs of giants who publish a lot of fantastic machine learning research and who paved the way. Just, once again, every single discipline in the world, I think, is that way. And you've got to give some props to everyone else who has come before. But we've really built a lot of our own algorithms internally. And then really only in the past 2 years or so, 2 to 3 years, have you been able to marry really powerful machine learning algorithms that if you were to try to run these algorithms on your home laptop, even a good one, It would probably take days to run.

Sure.

We can run all of this now in the cloud on incredibly powerful servers at a relatively speaking low cost, and we can do that in milliseconds and serve the results. So essentially it's a marriage of fantastic machine learning algorithms and extremely fast and scalable infrastructure. And when you merge those 2 together, You actually have a real-time machine learning system.

Okay, interesting. So how does it work? Like, okay, so I've been running, say, it's been like a couple months or a quarter, you know, I'm— you're obviously changing the content based on who I am on the site. Do you guys provide like kind of stats or like do I have like kind of a dashboard when I log in or kind of walk me through What do I kind of— can I tangibly see that you guys have kind of done for me?

Yeah, so we do have a dashboard, and you can absolutely track those 80+% improvements we're getting you. We actually make it easy though, if people don't want to necessarily believe our stats. You can also use Google Analytics or other third-party kind of analytics tools to see how well we're doing. And we run these A/B tests, and we prove it out really clearly for you right at the beginning. Like, look, within— let's say it takes 2 to 3 hours to install ListIgniter, and that's it. You're fully integrated in 2 to 3 hours.

Okay.

Within 2 to 3 weeks, you're going to be seeing on average 80+% improvements in your metrics. Mm-hmm. And so very, very quickly, It should be hopefully very clear that we're, we're doing our job. You know, don't believe me because like I'm telling a good story. We tell our customers, believe the numbers. We are a math and machine learning company, and we love— we want our customers to believe the quantified metric outcome.

Yeah, it's like data doesn't lie, right?

Data doesn't lie. Exactly right.

So yeah, no, that's That's really interesting. So, what's the cost for people to actually use the service?

Yeah, so it totally varies based on the size of the site, so how much traffic you have, how many SKUs or items you have on the site. It definitely varies a decent amount, but it's in the few thousand dollars per month range, let's say, if we're really averaging it out.

So,

it's not cheap. But at the same time, we believe every single one of our customers has an extremely clear ROI.

Sure. Well, and you guys already have some, like, really big names using your service, right? So, you know, when, like, I don't know if you can, if you want to mention any of them, but, like, just even just right on your homepage, you have a bunch of, like, huge names that everybody's heard of, right? So it's obviously working, right? And so Because those companies could pick whoever they want to use, right? At the end of the day, like, and if it wasn't working, they wouldn't be using you guys. Like, so it's, and if your success rate is, is that high, which is like totally fascinating to me. So I want to dive a bit deeper into kind of just for, for my own interest is like, how do you kind of No, like you kind of mentioned it a little bit. Like I searched for, you know, like a beach vacation and you start recommending me shorts, but like what kind of other things do you kind of start suggesting to me or, or kind of, you know, prompt me with?

So it, it really depends on the customer, right? If you're a news site, we're gonna be suggesting articles like, articles or political articles. If you're an e-commerce site which has a lot of t-shirts, we'll recommend other t-shirts.

Sure.

It really, really varies and depends. But I mean, I could add, it can be anything from articles to videos to music to images to items to buy, you name it. Our view is that every single potential place you can make a recommendation, which is truly almost everything that you see, Online, everything is a recommendation. That recommendation should be optimized using machine learning. And here's sort of, I think, the kicker, at least we truly believe this, which is if you are not using machine learning in personalization on your digital property in the next 5 years, you essentially will cease to exist as a business.

Interesting.

You will not be able to compete. You just won't be able to compete. Because Google, Facebook, Amazon, all the big guys are already doing it, so you're already losing ground to them. In the next 5 years, everybody else will hopefully be using LiftIgniter, but they'll be using something in this space because the results are too dramatic to ignore, and your users simply just won't come to your site, or they won't buy a lot because they know they'll find it somewhere else.

That makes a lot of sense. So how do you guys deal with Something like Google Home or Amazon Echo? Is it Echo? Alexa?

Well, I think Echo is the official product name. Alexa is sort of the moniker, but I think Echo is the official product.

Yeah, yeah. So how do you guys deal with those voice-activated stuff, or can you?

Great question. The short answer is we can. We don't do a lot of that today. Actually, we don't do any of that. Sorry. We don't do any of that today. Mostly because it's still really early in a lot of cases. Those systems are fairly closed. I mean, Alexa has lots of skills and things like that. And I think longer term, we'll probably start working with the people who are creating the skills to, you know, it's actually really critical in places like voice activation or things like that to have really intelligent recommendations.

Mm-hmm.

Because, you know, those are situations where it's really difficult to— you can't navigate things very well. So the quality of what the system is automatically recommending to you is even more important and more critical to the system for it to be valuable. It has to make an intelligent recommendation right off the bat. I'll give you one quick example, another example of why this is so important for people. I'll ask you, Kevin, a question, Kevin. How often do you go to page 2 of Google search results?

Never. Like, maybe once a year, maybe, right? Maybe.

Yep, you got it. So you've now just proved why really good recommendations are so critical, right? One, you never even have to go to page 2. 2, you always go to Google. You probably don't go to Bing very often. No. Despite Microsoft building a great product. You've been trained that Google is the place to go to get your answers, and you know you'll get it immediately. Totally. Imagine if every single site has the same ability.

Sure. Interesting. Yeah. Well, to your point, like we're going to be there in the next 3 to 5 years, right? Easily.

Easily. Well, we're already there, right? Google, Amazon, and Facebook already have a lot of our traffic. So we're already there. And it's just everybody else who either needs to start doing this or face extinction.

Yeah. Fair enough. No, that's really interesting because you're right. Like even on, Just like Android or iOS or just any of the Google products, they're constantly recommending stuff based on— in some ways, a lot of people think it's convenience, but convenience is basically like machine learning, correct?

Yeah. Well, that's one form of machine. There's lots of forms. Convenience in our world, convenience is about intelligent recommendation. Which is, which has to come from machine learning to do it at scale.

Got you. No, that, that makes a lot of sense. So you kind of touched on this a little bit earlier, but I'm kind of curious, did you guys kind of originally self-fund the first version? You mentioned you took some investors, or that's how you guys got introduced, you and your co-founders. So like you took some investment at some point, but like walk me through kind of the early days. Did you guys self-fund this? Did you just get investment right away? Or how did that kind of come to be to build version 1 of

Yeah, no, we were, we're, you know, pretty lucky, especially with kind of the backgrounds that we have, and really, Andrew Ng's background as an expert in this space.

Sure.

He has a PhD in machine learning from Princeton and went to the, you know, spoke at the number one machine learning conference in the world and a bunch of other things. And we got into what's a fairly well-known program, Y Combinator.

Right.

That you've probably heard of.

Yeah.

We got into Y Combinator, and very quickly thereafter, there were a large number of investors who were interested in putting money into the company and helping us achieve our next level of goals.

Sure. No, that makes a lot of sense. So, I'm curious to know, what's kind of your typical kind of day look like working at, like, LiftIgniter? Like, are you Kind of on the phone a lot? Are you emailing? Are you kind of building some stuff? Like, what are you kind of— what's a typical day? I'm just curious.

Yeah, no, it's a great question. The first answer is it's actually really— it's changed over time, as one would expect, especially when you're a founder. You know, you kind of do everything. Some days it's sales, some days it's marketing, some days it's legal. I mean, it varies. And as we've grown, And have more and more customers, and frankly, more and more people interested in what we do. My role from a day-to-day perspective, I mean, I would say now I'm on the phone at least half the day. Okay. I think today I've talked to 3 different potential customers already, and it's, you know, only— we started talking at 1 o'clock today, so 3 different customer calls already today. I sent out one contract, actually 2 contracts today as well, and had to modify those to make sure we had the specific things the customer wanted in there. You know, now having this sort of discussion with you, which I would bucket more in the marketing side of things, right? Sure. And then have another custom customer site visit later today. So a lot of customer stuff. But, you know, it really like— I have emails about accounting things in my email account right now that I need to handle. We're doing a really big partnership that's going to be announced in the next 2 weeks.

Sure.

That we're kind of trying to get the T's crossed and the I's dotted on that. So it totally varies, but the truth is in the end what it comes down to is just it's a lot. It's exciting. I wake up energized every day, but there are— it's a lot. I mean, I usually run calls or doing work from 8 AM till about 8 PM.

Sure.

Every single day, 5 days a week, and then I'm usually working for at least 1 full day on the weekend. So it's at least 6 days a week for about 12 hours every day.

Sure. Wow, that's a lot. But when you're in love with it, it doesn't really feel like work, so I get that. And I'm always kind of curious To know, you know, like obviously, like we covered your background earlier. I'm always kind of curious to see what people are kind of, what a typical day in the life of like their startup is, because it's just kind of fascinating to me. So, but sadly, Adam, we're coming to the end of the show. So let's close with mentioning where people can get more information about yourself and LiftIgniter online.

Yeah, no, absolutely. Go to LiftIgniter. L-I-F-T-I-G-N-I-T-E-R.com. And obviously, you can also find me at Adam Spector on LinkedIn.

Perfect, Adam. Well, I really appreciate you taking the time out of your day to be on the show and look forward to keeping in touch with you and have a good rest of your day.

Yeah, no, my pleasure. Thank you. And if there's ways I can help out other aspiring entrepreneurs or obviously answer questions about machine learning, anyone who's listening, feel free to reach out. Happy to be helpful.

Perfect, man. Well, thanks again for doing this, and we'll talk soon.

Thanks, Kevin.

Okay, bye.

Thanks for listening. The music for the show is done by Electric Mantra. You can check them out at electricmantra.com and keep them in the future.