Latest / NC Tweener Talks / Robbie Allen: Automated Insights and Infinia ML (Exits) -> Bionic Health and Triangle Tweener Fund

Transcript

Robbie Allen: Automated Insights and Infinia ML (Exits) -> Bionic Health and Triangle Tweener Fund · NC Tweener Talks

• 00:00:04 - Robbie Allen But I was able to raise the million dollars and in fact I signed a term sheet for a million dollars on my 13 year anniversary at Cisco, which is when I submitted my resignation. So I had a lot going on that day and but then I was kind of off to the race. Welcome to Triangle Tweener Talks, a weekly podcast by Builders for Builders where we explore the startup journey from the idea to the exit and all the less between. With an exclusive focus on founders from the Triangle region of North Carolina. Tweener Talks is produced by Earfluence. Now here is your host, serial founder and general partner of the Triangle Tweener Fund, Scott Wingo. 00:00:44 - Scot Wingo Welcome to this episode of Triangle Tweener Talks featuring Robby Allen, serial entrepreneur and my partner at the Triangle Tweener Fund. This episode is sponsored by Robinson Bradshaw, a full service business law firm with a passion for supporting the Triangle entrepreneurial system bank of America. Their transformative technology group helps game changing tech businesses and founders realize their boldest ambitions across a wide range of technology sectors and smashing boxes. A Durham based lean design centric design transformation company. And special thanks to our friends at Earfluence for helping produce this podcast. I first met Robby over 15 years ago while he was at Cisco and as a side hustle he had built a generative AI application. That was 15 years ago. This shows you how far ahead of the curve Robby is. Like many Triangle founders, Robby comes from humble beginnings. He's from Greensboro, North Carolina and one of the first folks in his family to go to university. In this episode we hear his journey from the University of South Carolina, home of the Fighting Gamecocks, to MIT and how he quickly became a distinguished engineer at Cisco. One of the highest accolades an engineer can achieve. We then hear how he pivoted from working at a big company like Cisco to founding companies Stat Sheet, Automated Insights, Infinea ML and his latest startup, Bionic Health. That's four startups in a very short period of time. I hope you enjoy this interview as much as I. I do. And now here is Robbie Allen. 00:02:11 - Robbie Allen So I was born in Burlington, North Carolina which is about 40 minutes west of the Triangle. My dad worked for AT&T in manufacturing and my mom worked at Haynes Hosiery in Winston Salem. And so a very kind of decidedly manufacturing oriented household. I like to say I don't think I heard the word entrepreneur until I was in college. So I didn't think that starting companies was something that was available to me at the time. I thought you go and hopefully get into College. I was the first one in my family to go to college. And then the whole point of that was to get a job. And so anyway, growing up, you know, just kind of a typical childhood, we moved around a fair amount because my dad got transferred and we lived in Winston Salem. We got transferred to Oklahoma during my middle school years. And then my 11th grade year, they decided to move back to Burlington, North Carolina, because being small town folk, they thought wherever Robby graduates high school is where he will live the rest of his life. Because that's kind of what they did. And so they thought, crap, if he graduates high school in Oklahoma, he's going to live in Oklahoma the rest of his life. So move me back my 11th grade year, I wasn't very happy about that. But one thing that happened my 10th grade year was I took my first class having to do with computers. It was called Business Machines. And you know, I didn't know anything about it, but it was one of those kind of magical moments in my life. Pretty early on when I was in the class, I was like, I want to do this rest of my life. And so I'm kind of one of the lucky few that got exposed to something earlier in their life and then realized that's what I wanted to do. And so I knew from 10th grade on I wanted to be in computer science. 00:03:54 - Scot Wingo Yeah, that was like the IBM PC era, right? They just come out. 00:03:57 - Robbie Allen Yeah, that was 1992. 00:03:59 - Scot Wingo Yeah, yeah. So was that a. An Amiga? Was it an IBM PC? Just a vanilla, vanilla. Kind of goes right at Windows, maybe. Windows was 1991. 00:04:08 - Robbie Allen Yeah, yeah, yeah. And they had. It was IBM business machines, I guess, is what that was about. And so did that, you know, enjoyed it. And then I knew I was going to study computer science in college and but just to kind of relate. I wasn't very happy with being moved back to North Carolina because again, I was middle of high school. And so I applied to two colleges, unc, which I was a lifelong UNC Tar Hill fan. I like to say I came out of the womb of a Tar Hill and I applied to University of South Carolina and I got into both. You know, back then it was a lot easier or they weren't as discriminating as they are now. I got into both. And instead of picking the place that I loved growing up, it was only 20 minutes from home, I decided to pick the place that was much farther from home so I could get out of Dodge. And so I picked University of South Carolina, which is another thing we have in common. And I went there for my first couple of semesters and really liked it. 00:05:06 - Scot Wingo Yeah. And then so did you, after you took that 10th grade class, did you learn to code kind of on your own or you really didn't do that until you got to college or did you take like a high school class or something? 00:05:15 - Robbie Allen Yeah, there weren't many, you know, back in 92, 93. There weren't a lot of classes. I don't even remember. You know, oftentimes there's computer. Ap, Computer science. I don't even think that was available in my schools. That might have been like the only computer related class available. And so it was one of those things I was interested. We. I got a computer at home, but I was mostly just kind of hacking on it. You know, I was for fun. Like remember the old floppy disk and, you know, all the different types of software you could get. I would go and make copies of those and, you know, try to hack the computer without really knowing anything about programming. That didn't come until I was in college. 00:05:50 - Scot Wingo Yeah. Okay, cool. So you go to usc, you're in comp sci. 00:05:53 - Robbie Allen Yep. 00:05:54 - Scot Wingo At that point, for a long time, usc, the comp Sci was in the math department and not engineering. Was. It was still separated that. 00:05:59 - Robbie Allen Still separated, still. And that was not too uncommon in general back then that Matt, you know, computer science was kind of a afterthought in the math department. 00:06:06 - Scot Wingo Yeah, yeah. So you did a couple years there. 00:06:09 - Robbie Allen Yep. And so my high school sweetheart, we met at. In Burlington, she went to ecu and so we were kind of doing this long distance thing. She was at ecu, I was at usc. And then after a couple years I was like, you know, why don't we just get back together and, you know, you could either come to USC or I could go to ecu. We picked ecu and so I ended up transferring to ECU into their computer science department. 00:06:35 - Scot Wingo Okay, so that's school number two. 00:06:37 - Robbie Allen Yep. And I have a long list of schools I've been to. Turns out I really like the college environment. Yes, sir. 00:06:43 - Scot Wingo Gamecock and a Pirate. So that's a good combo. Yeah, yeah. 00:06:46 - Robbie Allen And then. 00:06:46 - Scot Wingo But the ECU comp Sci program at that point was probably a bit of a step down from South Carolina, I would imagine it was. 00:06:52 - Robbie Allen Although, you know, the funny thing is I've been to, you know, I often like to tell especially sort of high school students that I've been to everything from ECU to mit. And some of the best education I ever got was at ecu and some of the worst was at mit, mostly because schools like ecu, they're not. Especially ones that aren't known for, like, deep, heavy research. They're there for teaching. And so I go there, and the professors were all into teaching. That's all they cared about. And so the quality of the education was much higher than what you'd get in some other classes where the professors may be a noted researcher, scientist, but they didn't really care at all about the teaching. And so it turns out, for me, the ECU computer science department was awesome. It was still kind of very similar in terms of. It was just kind of starting out and kind of rising up above electrical engineering and math. But it was the kind of thing that, you know, the professors there were great. You know, they really. It was a pretty small environment, so I could, you know, get the classes I'd want and get the professors I wanted. 00:07:52 - Scot Wingo Yeah. And you meet a professor. Like, I had a professor at USC that just let me do whatever I wanted when he found out I was kind of interested in stuff. Yeah, he's like, oh, here's like, you know, here's a log into this system. Go crazy. Wow. Okay. 00:08:02 - Robbie Allen Yeah, yeah. So I was the. I was the system administrator for the computer science lab, you know, so, yeah, I had access to all the systems. It was great. 00:08:09 - Scot Wingo Yeah. Cool. Now, did you graduate from there or you're gonna. You hopped again? 00:08:13 - Robbie Allen Nope. So I started. When I went, moved to ecu, I decided I wanted to start doing internships because, again, in my mind, the whole goal of going to college was to get a job. And so I did an internship at IBM in 1996. And this is kind of full circle, funny situation. So it was at building 800 at the front, what's now known as the Frontier. That's where IBM networking hardware division was based, which was their sort of old router and hardware business that competed with Cisco. And so anyway, I did an internship there, and that was like my sophomore year, and it went well. And I heard about this company down the road called Cisco, which is literally just three miles, four miles away. And they were apparently eating IBM's lunch when it comes to routers and switches and that type of thing. I was like, huh, that's kind of interesting. And so maybe next summer I'll do an internship at Cisco. And so, sure enough, another year at ecu, and then I went and did an internship at Cisco in 1997. And that was a crazy experience because Cisco was. It was really kind of their heyday. And at the end of the internship, they took me aside and at this point I'm all in on programming. Like I'm just living and breathing that I'm, you know, at Cisco I was sleeping under my desk a couple nights a week. I was, you know, just spending all amount of time doing it. 00:09:30 - Scot Wingo But that network stuff, you're pretty close to the metal. You're like maybe assembly, maybe C. You're not like much above that. Right. 00:09:35 - Robbie Allen I did all of it. Like I would build websites internally for their intranet. You know, I would do networking programming as well. So I was kind of exposed to all of it. And so at the end of that they approached me and said, this thing called the dot com boom is happening in Silicon Valley and we need programmers as fast as we can find them. And so we'd love for you to we just ship you out to California and we don't care that you don't have your degree and you could just work on your degree in your spare time. How does that sound? And so again, I thought, well, the whole point of going to college is to get a good job. And it's like, all right, they're saying I have a good job and I don't even need to finish my degree, I'll just do that. And so I moved out to San Jose, California and I lived out there during the boom and the bust, which was just a phenomenal time. Again, this was sort of Cisco's heyday. And again, at this point I still hadn't really contemplated ever starting a company. I didn't think that that was something. I thought, I'm going to be a Cisco for my whole career. Let's talk. My dad worked at AT&T for 30 years. 00:10:31 - Scot Wingo Going from Burlington to San Jose is a big leap in culture. 00:10:35 - Robbie Allen It is. And you know, being in California, especially for a, you know, I was 21 years old or whatever, Northern California is just great. You know, there's so much to do, so many things to see. So I was just eating all that up. 00:10:47 - Scot Wingo Yep. And then did you actually like buy real estate or did you just rent out there? 00:10:51 - Robbie Allen So I ended up my manager at the time, I ended up moving in with him and we bought a condo together. And this was, you know, again like 2000 or something like that. And now at that time, real estate was doubling like every six to nine months. And so like I didn't have that much money and I went in and, you know, paid what I could for this condo. And then of course, you know, a couple years later, it had tripled in value. And so, yeah, I was just able to get into something then, which is great. 00:11:21 - Scot Wingo If only you had kept it, I. 00:11:22 - Robbie Allen Know I wouldn't be here today. 00:11:24 - Scot Wingo Could have been your best investment. 00:11:25 - Robbie Allen That's right. 00:11:27 - Scot Wingo Cool. So then you're out there, you're through the boom and bust. And I remember the bust was pretty, pretty, you know, fast and scary. There was like the Kleiner Perkins rip. Good times kind of came out. And then, you know, it just like was down, straight down. And Cisco got hit very hard because they were, they were giving customers loans to buy equipment. There was a lot of kind of like, you know, stuff that you wouldn't do today. Kind of questionable business activity from a lot of, you know, the Enron was like, you know, borderline criminal. But even Cisco and those kinds of companies ended up getting really shellacked during that time frame. 00:12:01 - Robbie Allen Yeah. And I remember, you know, driving around the Bay Area is notoriously pretty gnarly. There's a lot of traffic. And I remember one day driving into work, I'm like, you know what? There's not as much traffic today. What's going on? And it was the bus that started. And in fact, you know, Cisco had laid off a ton of people and therefore like a year, the traffic wasn't that bad. And initially that kind of struck me. I was like, oh, I get it now. There's not as many people going to work because they were all laid off. 00:12:24 - Scot Wingo Yeah. 00:12:25 - Robbie Allen And so, yeah, it was kind of a freaky time. And that's when I decided, I was like, you know what? I don't know if staying around here is. I mean, it's a high stress environment. And I thought, well, you know, why don't I move back to North Carolina? You know, I have family here. My high school sweetheart had come out to California with me and she, you know, we're from Burlington. So, you know, it kind of made sense for us to move back. And Cisco is very easy when it comes to, you know, working from home or you could work from any of their major locations. And so, you know, we decided kind of right around the time that the bust was happening that that would be a good time to move back. 00:12:56 - Scot Wingo Yeah. And were you, I guess you were never at risk of layoff or anything? You just kind of. 00:13:00 - Robbie Allen No, I was, you know, continuing to rise through the ranks pretty quickly again. I was one of those that was sleeping under my desk several nights a week. And, you know, I'd like to say I got a lot out of Cisco. Cisco was great for me. But I think Cisco got A lot out of me too. You know, I was putting in lots of hours and time and doing some good stuff. So. 00:13:16 - Scot Wingo Yeah, and it wasn't. That was kind of not the norm at Cisco. It wasn't a startup at that point. 00:13:19 - Robbie Allen No, no, yeah, yeah. 00:13:20 - Scot Wingo Was that this the John Chambers era? 00:13:22 - Robbie Allen It was, yeah. 00:13:23 - Scot Wingo Did you ever get to meet him? 00:13:24 - Robbie Allen I did, but not then. It was only till later. We were speaking at a conference together. This was a couple years ago. And I got to meet him. I got, you know, Cisco used to give out these little plaques, you know, for employee of the quarter or whatever, little awards. And I, you know, I got a few of those and so I took one of them with me because I knew he was going to be speaking at this conference and I would get a chance to say hi and so I took one and had him sign it, you know, and it was like from, you know, when I was 22 years old or whatever. And so that was pretty fun. 00:13:50 - Scot Wingo Yeah, yeah, yeah. Do you pretend to remember? He's like, but he was great. 00:13:55 - Robbie Allen You know, he was such an all star back then. Everybody kind of looked up to him. I always thought he was a great leader in general. 00:14:01 - Scot Wingo Yeah. And then so you moved back here and you're still at Cisco and that's like off Kit Creek kind of. Were they still, were they there at that point? 00:14:07 - Robbie Allen They were, yeah. They had been there, you know, since even before I'd started. They'd been there a long time and. But, you know, something else had kind of happened when I moved back. I decided that, you know, there was a lot of stock startups, obviously in the Bay Area. And I got to see that happen. A lot of them start and fold and I started to think, you know what, I got to meet a lot of entrepreneurs and I was like, I don't know, I think maybe I could do that too. I don't know if that's something that, like I said, is unattainable for me. And I started advising other startups. And so that kind of got this bug in my head a couple of things. One is I thought that starting a startup was something I could potentially do. And then also I started to really see the warts of being at a big company. You know, we were so big at that point, you know, 90,000 employees, especially after, you know, things started to come back. And my calendar was just wall to wall meetings every day. And it was just painful and, you know, just lots and lots of project managers scheduling meetings and then you have people that are just there to manage the meetings that are being scheduled. And it was just like totally unproductive. So I became pretty cynical about the effectiveness of big companies in general. And so that was the other thing that was kind of eating at me. He's like, I think I could do it better if I could start my own thing. 00:15:15 - Scot Wingo Yeah, sometimes. And then also Cisco started to kind of lose their way a bit. Right. So they missed the whole software defined networking thing by a lot. And then they had to acquire all their innovation suddenly leaned into acquiring things like they couldn't self innovate, which is kind of a sign that they're kind of trapped in the innovator's dilemma. And then they bought like WebEx, which didn't make, I mean, knowing the pandemic was coming, it would have been genius. But that wasn't really, it was just like a reason to sell more networking, like trying to create demand for networking. I don't know, didn't make sense to me from an outsider's view. 00:15:43 - Robbie Allen Well, once you get to that size as, you know, like in order to Apple and all the Microsoft, all these companies have this problem. Like once you go to a certain size, in order to kind of get to the next level, you can't just have a billion dollar business. Like you got to have a multibillion dollar business. Right. And so the bar is so high for them to add meaningful revenue to their business that they have to start doing crazy stuff. And again, this is where, you know, these big companies I think act unnaturally in most cases once they get to that size because like it's just almost impossible, maybe not possible except in just a few rare cases or something like AI comes along that kind of boosts your or juices your ability to innovate that, you know, they're just out searching. It's like, all right, WebEx, we're going to do telepresence, we're going to do all these things, you know, just to hope that something hits that we can get to our next level of revenue. 00:16:31 - Scot Wingo Yeah, okay, so you're sitting there enjoying the Cisco life to some degree, but you're seeing the cracks in the wall. Then, then what happened? 00:16:39 - Robbie Allen Well, as any good engineer that's interested in startups would want to do, I thought, you know, if I'm going to go start a company, I need to go learn how to be an entrepreneur. And so how do you do that? Well, you go take a course, you go, you know, read the book. And not knowing, of course, that the best way to learn about being an entrepreneur is by doing, not by reading a book. And so. But I didn't know that. Again, I'm an engineer and so we go and read books for things. And so I thought, well, what I'll go do is take, you know, maybe what I'll do is go back to school. Because that's also another answer that I typically kind of resort to in a lot of cases. And so. But at that point, I still hadn't got my undergraduate degree. You know, it had been, I don't know, eight years in. 00:17:18 - Scot Wingo Did that bother you or you didn't care? 00:17:20 - Robbie Allen Didn't bother me. I didn't really care because like, and I knew the classes I needed to take weren't like classes that are meaningful. 00:17:25 - Scot Wingo It was, you know, whatever, five or six. It wasn't like Killian either. 00:17:28 - Robbie Allen Yeah, that's it. And so I thought, ah, it sucks. Well, another thing had happened during this time I had started writing books. And so for. I got to know the folks at O'Reilly Media and I started kind of editing books for them. And then I named a co author for a book and I liked it. And in fact what I found was that's a really good way to learn about a topic is to write about it. And so at Cisco, it's kind of this virtuous cycle where they would say, okay, Robby, you know, we need to redeploy our DHCP infrastructure. Which is, it's the way that IP addresses get allocated to devices for those that aren't familiar, which I'm sure everybody's familiar with dhcp, right? 00:18:07 - Scot Wingo Yeah. You see the field and you're like, oh, that's scary. I know. I don't want to mess with it. If I have to mess with it, stuff's going down. 00:18:14 - Robbie Allen And so. Or they say, robby, you know, we need to redeploy DNS throughout all of Cisco's infrastructure. You know, we want you to be point for that. And I'd be like, I don't know that much about DNS. And so again, what I would do is go and spend a bunch of time researching it. But then I found if I started writing about it, that's a good forcing function to really learn about it. And so what would happen is typically it would coincide, you know, with Cisco saying, we wanted me, they wanted me to do a project is then I would go to O'Reilly and say, hey, you don't have a book on this topic. How about I write one for you? 00:18:41 - Scot Wingo Yeah, and I don't have any O'Reilly's famous for having fun animals on the covers. Did you get to pick your animals? 00:18:46 - Robbie Allen You don't get to pick. To pick. 00:18:48 - Scot Wingo What were the animals on your book? 00:18:49 - Robbie Allen So, you know, So I wrote 10 books and so I got, holy cow, there's a baboon, there's a tuna. There's, you know, all different kinds of things. There's a cat that was on the active directory book. 00:18:59 - Scot Wingo So you wrote, you just casually wrote 10 books? 00:19:02 - Robbie Allen Okay, yeah, yeah. So, you know, 10 books in about five years. And so I was just cranking those out and again, that was kind of filling my. So as I, I was working 80 hour weeks at Cisco, but then eventually I was kind of like, I don't know if it's worth spending 80 hours a week at Cisco. So I started trying to find other things to fill my time. And so writing books kind of was like my, you know, 20 to 30 hours a week and then Cisco would get 50 to 60. 00:19:23 - Scot Wingo Yeah. Have you ever done the hourly profit on writing a book? 00:19:28 - Robbie Allen No, no, no, you would never write a book. 00:19:29 - Scot Wingo It's like $3, maybe worse. 00:19:33 - Robbie Allen And then there's the opportunity cost with your time and all that. 00:19:35 - Scot Wingo Yeah, it's really good for branding though. 00:19:37 - Robbie Allen Yeah. And then I started, you know, I started going on the conference circuit and started speaking a bunch of technical conferences and doing all that. And so that was all good and well, but like again, at that point I still hadn't had my undergraduate degree and now I want to go back to school and get my master's. But one thing is, most programs kind of require an undergraduate degree for you get your master. So I went to UNC and Duke and asked them, hey, can I, would you accept me into your master's program if I don't have an undergraduate? Now I've done all these other things, I've written books, I've done these things. Speak at conferences, would you allow it? And it sounded like that they were open to that, that there were, they could make an exception, that if I just, you know, I was really close to finishing it, the degree. But you know, because I've done these other things, maybe they'd let me happen, let me do that. And then my wife, you know, one day pulled the kid card, so we hadn't had kids at that point. And they said, okay, Rob, you're going to get all this way. You got like four classes to finish your undergraduate and then you're not going to complete it, you're going to quit. And I'm like, oh gosh. Pulled the kid card on Me, that's not fair. And I was like, ah, you're kind of right. You know, if I. If I finished my undergraduate, then I could just apply to anywhere. I could go anywhere I wanted, really, or I could apply to anywhere. And so I decided to do that. And so I, you know, I talked to ecu, they said, okay, you know, I could transfer to another place. And by the way, in intervening years, I went to. I took classes at NC State, I took classes at San Jose State, took classes at Stanford. So I. 00:21:03 - Scot Wingo To chew away at the undergrad piece. Okay, so you weren't like a full year behind, maybe half a year or something by that point. 00:21:08 - Robbie Allen And then with ecu, like, I could transfer them all to unc, but again, I'd lose some credits and it would take longer. And I was like, well, I could just buckle down and take these four classes at ecu. Most of it would be remote. I'd only have to go there on occasion, so I'll do that. And so that took like a year. And I went ahead and knocked out the undergraduate degree. The great thing about that is, again, I can now apply to any college. Right. Because I don't have that. Have to get an exclusion. 00:21:31 - Scot Wingo So you're officially a pilot. 00:21:33 - Robbie Allen I am. That is one of my affiliations. 00:21:35 - Scot Wingo And now you love pirate basketball more than UNC basketball? 00:21:38 - Robbie Allen No, that's never changed. I've always been pro UNC when it comes to sports. And so as a computer science person, one of the places I'd always kind of aspire to go to is mit. And so I applied to mit, they accepted me, and so I talked to Cisco, and they let me work part time while attending MIT for a couple of years. And so was this a master's or. 00:22:03 - Scot Wingo Like a master's mastership? 00:22:04 - Robbie Allen Yeah, it was in. They had a. I wanted to do kind of a joint thing with the engineering program and the business school. And so they have an engineering management degree. That was essentially that. And so I did that. It's called System Design and Management. SDM was the name, is the name of the program. Highly recommended. And so I did that and commuted between here and Boston for a couple of years and really enjoyed that. That was 2004, which is right when the Red Sox started winning world championships, which was fun and just had a great time. And, you know, it was pretty early on, I took. I think it was called New Enterprises, was like one of their first entrepreneurship courses. And I was like, oh, gosh, why? I don't. I don't know why I thought I needed to take this course to learn how to be an entrepreneur. It didn't work, but anyway, so I did that. You know, it was an amazing experience for a few reasons. One is they would bring in people to speak at lunch and they had all these different, you know, just amazing people give talks. You know, one that was noticeable was the guy who invented the Crest spin brush, which is one of the first, you know, economical spin brushes for toothbrushes. And he made $300 million by putting a toy motor in a toothbrush. And that was essentially his innovation. I was like, I don't know, you know, there's all these very expensive, you know, spin brushes, but what if I made one that was super cheap and he was a toy maker and he thought, well, I'll just put a little toy motor in and see what happens. And he sold it to Crest. And I was like, that's pretty cool. And then there are people like Peter Diamandis, who was an MIT alum that gave a talk. Mark Cuban gave a talk. So one of the things I heard pretty consistently from a lot of these entrepreneurs was that you really want to be passionate about the business that you're starting because the going is going to get tough and going to want to quit. And if you're not really excited about what you're working on, then it's going to be really easy for you to fold up shop. And that resonated with me because again, anytime I could get passionate about a topic, whether it's writing books or whatever I was doing at Cisco, you know, I'd spend whatever time necessary to be successful at it. And so, you know, that's a little bit controversial advice which we could get into. You know, it's, there's, you know, by definition, there's not that many things that you're passionate about that are great business opportunities, right? And so, you know, like my 10 year old son, you know, I love him, but like, you know, he's not a great business opportunity. You know, there's, you know, I like you, I like you at C basketball, but, you know, that's not a great business opportunity. Or maybe it is. And so that's kind of what I started with. You know, I'm gonna make a list of what are the things I'm passionate about. And the list consisted of three things. It was doing interesting things with data Web programming and UNC basketball. And so this was 2006, you know, just finished graduating from MIT. I ended up getting two master's degrees. I had taken enough courses at the end of it, they said, well, Robbie, if you just take three or four more classes, you can get a second master's. I was like, cool, yeah, I'll do that. 00:25:01 - Scot Wingo So it was like, what, Comp Sci and a business thing or something. 00:25:04 - Robbie Allen It turned out it was actually civil engineering. 00:25:07 - Scot Wingo Wow, so you can build a bridge too? 00:25:09 - Robbie Allen Yeah. I probably wouldn't put me in charge of that, but at least I have the degree. And so I came away and I was like, all right, I'm going to start a company now because I got this fancy degree and I should know how to do it. And so I made that list up and I thought, you know what, what would be interesting is no one had really done a lot of stuff with sports statistics on the Internet at that point. And I figured, you know, if I could get access to that data. There's all these questions I have about UNC basketball. Like, are we really. Do we really have, you know, typically play a team that goes off against us from the three point line, right. And they do better than average just against us. You know, they perform at their average for most games. But for some reason, we have outliers on a regular basis. 00:25:55 - Scot Wingo Are referees really biased? 00:25:56 - Robbie Allen Are referees really biased? Are, you know, are the preseason polls as really bad as we all think they are? You know, are there people that are perennially overrated and underrated in the polls, things like that that, you know, you couldn't find answers to? And if I had the data, I could answer those questions. And I figured, well, if I can, I have those questions about UNC basketball, I could build the website in such a way that for any team, you could answer similar questions. And so that was the beginning of stat sheet, which was the kind of very first thing that I built. And this was, again, I was still at Cisco. I came back to North Carolina full time. And at that point, once I had the degrees from mit, they made me a distinguished engineer, which was their top technical position. And it's great because it's kind of like a tenured position at a company. I mean, you know, not to say that you're bulletproof, but like, you can get away and you can. You have a lot of job safety if you're a distinguished engineer. 00:26:51 - Scot Wingo Yeah, probably get to do a lot of big thinking around architecture, not. Not as much doing, not as much sleeping under the desk. 00:26:56 - Robbie Allen And they bring you in. You know, this is kind of another story I'd love to tell is that, you know, every engineer, when you start, your goal is to be. I mean, if you're ambitious is to be a distinguished engineer. Right? Like they're the, At Cisco, they were the cream of the crop. A lot of the people that were distinguished engineers and fellows were the people that wrote the protocols behind the Internet. Like, you know, and so just huge impact. And so you kind of aspire to be that. And so one day, you know, they brought me into, my manager brought me in office, said, robby, congrats, you're a distinguished engineer. And I'm like, oh, that's awesome. Something I've been working toward 10 years, you know, what does this mean? He's like, oh, just come into work tomorrow. I'm like, oh, okay. But like, I don't know, you know, there's no ceremony. Yeah, there's no ceremony or there's no. 00:27:44 - Scot Wingo Watch, there's no nothing. 00:27:45 - Robbie Allen Yeah. What do we, you know, do I get up on stage Something. 00:27:48 - Scot Wingo Yeah. 00:27:49 - Robbie Allen And so, you know, is this. That was a realization of like the 10 years it took me to attain that goal. What was way more fulfilling was the 10 years not the point at which you actually achieved a goal. And so that was informative for me in that, you know, any company I ever started after that, you know, a key value is enjoy the journey, don't wait or hope that you're just, you know, going to have some big exit or you're going to have some amazing end result. Because that's pretty fleeting actually. The thing you need is to have a sustainable and enjoyable path to get there. And that was the thing even for me. Again, I was super fortunate to get to that position at Cisco, but like, I just had so much fun along the way that was way more important. So anyway, I started Stat Sheet kind of nights and weekends. And so again, I have this kind of spare 30 hours. I was tired of writing books at that point and so I thought, well, I'll start a company. And so I started Stat Sheet and you know, worked on that for two years or so and it got to become pretty popular again. This was 2007 and 8. I started adding more sports. I started with college basketball, but then I started getting data for college football in the NFL, NBA, baseball. 00:28:57 - Scot Wingo You probably had to go scrape the data, I imagine. Like, did you have a pretty sophisticated web crawler for this data at that point? 00:29:02 - Robbie Allen So it's a couple things. One is some data there's no providers for especially like historical data and that type of thing, but for like the standard box score and especially any kind of real time data. Actually I started from day one going and getting it licensed because my thinking was, you know, if I build this up and someday it's very successful and somebody wants to acquire it, if I say, oh, all the data, you know, I scraped from somewhere else that could be viewed negatively and may impact my ability to partner with other people or get acquired. And so I was like, you know what, I'll just bite the bullet and do that from the beginning. And then it also helped that I applied for an NC Idea grant and I won. And that was like in year two, 2009, was it 50k back then? 50k? 00:29:42 - Scot Wingo Yeah, still 50k. 00:29:44 - Robbie Allen And the funny thing is, you know, I applied the year before that and this was back when it was Word documents that you fill out and I misread the limitations. It said like, you know, only, you know, max of 300 words or 300 characters. And I did words or I did, you know, whatever. And so I blew out the limitations and they automatically disqualified me because they said it was too long. 00:30:05 - Scot Wingo You're a writer, you're going to write. 00:30:06 - Robbie Allen I know. And so they said, well, just plug in next year and make sure you follow the rules. And I was like, okay. And so then the next year I was able to win, which I always, you know, I thought was amazing. I was, I've always been a huge fan of the NCI data grant program because like how many states can you get a grant? You know, that is, they don't even take equity and they're going to give you $50,000 non dilutive. To me that was a great deal. So anyway, I did that, that kind of paid for my second year of operations with paying for stats and hosting and that kind of thing. But then it's pretty obvious to me that I was all in on statsheet. I really enjoyed doing it. My interest in Cisco was kind of waning. I was just totally done with the big company politics. 00:30:46 - Scot Wingo What did they think at Cisco when you were working on this? They think it was crazy or they. 00:30:49 - Robbie Allen Well, they didn't really know too much. I mean, I told my manager at one point I had a sports stats website I was maintaining, but a lot of engineers would have little side projects like that. So it wasn't a big deal. But again, it was mostly something I did outside of the purview of Cisco. It wasn't impacting my job, but I was interested in enough. I was like, I need to move and just focus on that. And so I decided if I could go raise some money, then I would leave Cisco and focus on statsheet full time. But I Was not completely sold on it. And so I did what a lot of entrepreneurs or budding entrepreneurs do is to go around and try to talk to some people to get advice. And then that's when I came across this guy named Scott Wingo. Again, this was probably 2007 or 8 when we first met. 00:31:38 - Scot Wingo And, you know, 2008, I was probably depressed. In 2008, I would not have been a positive entrepreneur. 00:31:42 - Robbie Allen Is that right? 00:31:43 - Scot Wingo The great. The great. Towards the end of it was like, the great financial crisis was upon us. 00:31:47 - Robbie Allen That's right. I think it was before that. 00:31:49 - Scot Wingo Yeah. 00:31:50 - Robbie Allen Because you were very positive. I was like, you know, hey, I think we met at a Starbucks, of course. And it was, you know, I was. I told you the story, and you're like, yeah, you should leave Cisco and just go start your thing. I'm like, I don't know. But, like, I got this great job and, like, who knows, you know, if statute will be successful. And you're like, oh, you'll figure it out. 00:32:06 - Scot Wingo And I'm like, we also had. You had received acquisition offers, if I recall, from like, you know, ESPN or EA or like, you know, and you were like. And then your traffic was like, your mr, Your maus were just, like, up and to the right. It was crazy. 00:32:21 - Robbie Allen Yeah, yeah, yeah. ESPN had made overtures and invited me up to Bristol. That was kind of a pretty fascinating time. I got to be at the ESPN desk and got my picture, you know, as if I was. 00:32:33 - Scot Wingo To me, it's like so obvious. Like, you had. You had lightning in a bottle, but you're like, yeah, I may just like, stay at the same time, like, most people die for this thing you've created, and you kind of have just like, you know, not stumbled upon it, but you figured out something. 00:32:45 - Robbie Allen Yeah, well, if you think about it, though, you know, I was very brought up, Risk adverse, Right. Like, you know, the idea of starting your own company again, the whole point was to go and get the stable job that you work at for 30 years and then take retirement. Right. And so for me to kind of deviate from that was a pretty big deal. But again, at that point, I was confident enough that I could. I'd rather make the bet on me. You know, a lot of people and I. I used to write a lot about this because, you know, people love to associate big company working at a big company with safety and, like, it's the less risky option compared to going and starting a company. And I've done both. And I always thought that staying in a big company is actually riskier on a few fronts. Like with a startup, at least you have some direct influence over your ability to control your situation at a big company. I mean, again, even though I was a distinguished engineer, I knew of other distinguished engineers that got laid off. Like if they, especially with Cisco at the time, occasionally they would lop whole departments and whole divisions. Right. And maybe in some cases they give you the chance to reapply internally for a position, but in other cases, they just will let go of the whole department and you have no control over that, no ability to influence. 00:33:52 - Scot Wingo We had people leave Channel Advisor to go to Cisco because it was more stable and then get laid off three months later. And then, you know. Yeah, and then try to come back. Yeah. 00:33:59 - Robbie Allen And then there's these other issues that, like, I, you know, often talk about that your network, your personal network atrophies over time. Your skill set atrophies over time. And again, you're in this kind of small little bubble. Typically, again, inside of a big company, you know, you have a small network of people. You're talking, you know, and you're kind of working on this kind of very specific thing. And so you're not out talking to people of different areas or different companies that often it's all kind of within the same thing. And so why it's so risky is, let's say you're there 15 years or 20 years, which I know a lot of people that are still at companies like that for 20 years, and then you get laid off, and then now look at your prospects there. Well, you better hope that whatever that little thing you were working on that exists at some other company, because if it doesn't, it's not exactly transferable to many other places. And then your own network has also atrophied. You not been out, like, talking to people that maybe could potentially hire you at a different company. And so I think it really sets you up. While it may seem safer in the moment, I think in the long term it's actually much more risky. 00:35:05 - Scot Wingo I think this AI wave is a really good example of that because I'll go talk to people at big companies. They're like, what are you excited about? I was like, oh, my gosh, this AI stuff's crazy. I've been doing this. They're like, oh, we're not allowed to use it. And their mindset is they could use ChatGPT, like when it first came out, and then there was an edict not to. And that's kind of like when their knowledge two years ago of AI stopped and you know, maybe sometimes they're a little bit curious, but you know, it's hard when you have a job. You're not going to go, you know, you're unique in that way that you spend an extra 20 hours doing stuff. Not everyone does that. And you know, I think that's going to be, that's going to be the biggest skill gap that a lot of people at big companies have in this next, you know, three or four years. Interesting to see how they navigate that. 00:35:43 - Robbie Allen Yeah, I agree. 00:35:46 - Scot Wingo So you decide. So how much money do you, did you want to raise? Do you remember? 00:35:49 - Robbie Allen So one of the challenges, I'd written a bunch of code at that point for stat sheet and what I didn't want is to go raise some money and go hire one other engineer because like at that point, like I said I'd, you know, started adding a support for lots of other sports. You know, adding one engineer, you know, that wasn't going to move the needle for me that much. I needed to hire like at least three or four. 00:36:07 - Scot Wingo Yeah. 00:36:08 - Robbie Allen And so, you know, I kind of came up with the number. If I could raise a million dollars, you know, that would make it worthwhile for me to lead the safety of Cisco and really I could make a good effort at, you know, taking statute to the next level. And this was in 2010, back when a million dollar seed round was, you know, a pretty big number. But I was able to do it. You know, I found the right firm that was interested, specifically the partner at the firm who later became my coo, Scott, you know, great guy. He was also very into sports and you know, he's a dookie, I'm a UNC guy. And so we always had healthy banter about that. But I was able to raise the million dollars and in fact I signed a term sheet for a million dollars on my 13 year anniversary at Cisco, which is when I submitted my resignation. So I had a lot going on that day and. But then I was kind of off to the races. 00:37:01 - Scot Wingo Yeah. Was Cisco freaked out? Did they try to keep you or. 00:37:04 - Robbie Allen You know, I told them what I was doing and you know, of course they, you know, would make overtures and say that they want to keep me, but like there was just nothing they, literally nothing that they could do that would make me want to stay. 00:37:17 - Scot Wingo Yeah. So you're still stat sheet at this point and then you've raised $1 million and you start going at it. 00:37:23 - Robbie Allen Yep. And so we did that for a year, year and a half and you know, it's statue was a consumer Internet company. And especially back then, you know, typically how you build those to scale is that you really don't focus on revenue in the early days because you need a lot of eyeballs to be able to get to the point where you can have. You could turn on the spigots that would generate revenue. And so what you often would want to do then is not put barriers in place for adoption, because again, you're wanting to grow your user base as fast as possible. Because if you don't, if you can, you know, you have to get to tens of millions of users. If you don't get to tens of million users, you're not gonna be able to get enough revenue to make it worthwhile anyway. And so turning on revenue when you only have a hundred thousand users didn't make any sense. Again, to me, it's kind of crazy logic. I mean, that was the way that. 00:38:14 - Scot Wingo Today we call it PLG product led growth. 00:38:16 - Robbie Allen Yeah, but mine was even worse because the way that you monetize is through advertising. And so that made it even more challenging. I mean, it was kind of. Twitter was doing the same thing at the time. So you could kind of point at that and say, oh, that's the way we're going to do it is, you know, and. But the problem is, you know, sports statistics in general is not a huge segment of the Internet. A lot of people think sports, wow, that's such a big category when it comes to actually Internet viewing. It's not that big. And the biggest problem is it's pretty big if you aggregate all the sports together. But if you look at any one sport, it's just not that big. And it suffers from seasonality, that is, you know, college basketball is great essentially for the month of March. You know, outside of that, it just doesn't get nearly enough eyeballs to make it meaningful. 00:39:00 - Scot Wingo Yeah. 00:39:01 - Robbie Allen And that tends to be the way it works around most playoffs for those sports. And so what happened was I had to start supporting lots of other sports so I could fill in those gaps of seasonality throughout the year. So anyway, that kind of reared its ugly head when we're trying to figure out, all right, how, you know, we raised $1 million, now we're going to have to raise a Series A, and we're going to need to at least be able to tell a good story about how we're going to get to some sort of revenue. And it just was looking like that was going to be a hard pitch. And so what we decided kind of at the point of raising our Series A was, all right, we're going to actually stop doing the consumer piece, and we're going to switch and take this technology we built, which we haven't talked about yet, and turn it into a B2B company. And again, this was pretty unusual at the time that at the point we're raising our Series A, we, you know, changed the mode in which we were going to try to get revenue that is from a consumer to B2B. And we even changed the brand of the company. Went from stat sheet to automated insights. 00:39:57 - Scot Wingo Yep. 00:39:58 - Robbie Allen And really, the core technology that I had built at that point was this notion that if I had all this data, I could automatically generate narrative out of it. And so the way I would talk about it is I wanted to compete with espn. And espn, if you look at them, they really have two main assets. They have video rights, which they pay billions and billions of dollars for. I can't compete with that. But then they have all these personalities that write content and that I thought, well, I could possibly compete with that. In fact, I could probably have a lot broader coverage than they do if I had access to all the data, because then I can write about every team, just not the most popular ones. So I could have the long tail. If you remember back then, long tail was a very hot kind of mode in which people were trying to say, well, you know, yeah, you know, you got all the people focused on the big head, but if you looked at the long tail and you were able to get to that, you know, that would maybe even be bigger than just the head. And so I thought, well, I could do that for sports. And so anyway, developed this technology that would take in, say, a box score of a game and then look at a lot of historical data and then generate a recap of that game automatically. And it was really the notion that if you had all this data, I can summarize the data in narrative form. And so that was the thought that what we could do is just take that core idea and apply it to business, because lots of businesses had data. In fact, the era of big data, which was a buzzword at the time, was really starting to take hold. But people didn't know what to do with all this data. They were told, you need to collect all this data. And so they started collecting it. But then eventually they were like, so why am I collecting all this data? What am I going to do with it? 00:41:30 - Scot Wingo And maybe they put up a dashboard. But a dashboard is not super interesting. It's not actionable and it's. 00:41:36 - Robbie Allen The problem is with the dashboard is it's very passive. Like you have to go to it, you have to decipher the meaning. I often say that dashboards require you to be a data scientist at a certain level and at least for executives, most executives aren't data scientists. And so wouldn't you instead just rather have a sentence that said, hey, your sales are down 33%, but don't worry, it was the same way Q3 of last year versus you having to look at the chart and say, okay, actually I see the trend and this is what's coming out of it. It. And so that was the bet is like, all right, let's take this technology. We refer to it as natural language generation. That's kind of the computer science Y term. There's natural language processing, nlp, which has been around for a long time, but then this kind of more emerging field called natural language generation or NLG. Now we call it generative AI. But essentially what we were starting in 2011 at that time was one of the first commercial generative AI companies. 00:42:28 - Scot Wingo But you weren't using an LLM per se. But you were. Yeah, yeah, you were. You know what was interesting though? And ultimately I was on your board, so I know a lot about this story too. So, you know what was interesting is every article is unique and you could personalize it to the person, right? 00:42:42 - Robbie Allen Yeah. 00:42:42 - Scot Wingo So if Robby was going to get an update, it would be largely Tar Hill basketball and then maybe like some other things going on of interest, like maybe in the ACC or something. Whereas, you know, someone else that's interested in the sec, they would get very customers kind of thing. So it was very unique in that way. The other thing you did that was really smart was and I don't know if this was intentional or not, but then, you know, the journalists in the world started to realize you were creating robot writers. So there's a lot of really interesting dust ups around. You know, are you going to put all the journalists out of out of work? And you know, is this ethical and things? And you know, part of being a startup that's hard from a marketing perspective is just getting attention. And if you can do something kind of controversial, it's an attention magnet. I don't know if you enjoyed the attention at that point. 00:43:26 - Robbie Allen We did. In fact, it was leaned into it. 00:43:28 - Scot Wingo At one point and started writing about it pretty aggressively and joking about people, things and things. 00:43:33 - Robbie Allen It was awesome. What I found was, you know, journalists were the ones that are in charge of what articles are being written, and if you have something that they view as potentially a threat to their job, then they were happy to write about it. And so we got covered in like the Thanksgiving edition of New York Times one year we got Wall Street Journal just like a lot of early stories where it was kind of a robot hand holding a pen, you know, that was. For whatever reason, they really gravitated toward that. And so that was one of the things we had going for us in spades as Automated insights was like we'd have all this press coverage pretty easy. 00:44:07 - Scot Wingo Yeah. And then did you shut down stat sheet altogether at that point or it still kind of just did its thing? 00:44:12 - Robbie Allen Well, this was kind of a controversial thing internally. So even though I had done, I probably wrote 90% of Statsheet at that point, I was in favor of just shutting the whole thing down because I thought it was a distraction. But it's one of these things, and this is one of the hard things about a startup is if you have something that's just kind of like a decent amount of success, it's really hard to turn it off because you're like, I don't know, some people say they really like it and so we're really going to turn off something that people like. And the answer is, yeah, if it's kind of not core to your business and it's distracting you from the thing that you should be focused on, then you should shut it down. But it was super hard for us to do that internally because, you know, for years that's what we were known as. That's what we had developed and what we had built. And again, around these playoff parts of the year, you know, whether it's March Madness or whatever, we get this huge influx of traffic. And so I was pretty much in favor of shutting it down, but, you know, I lost that bet for a few years just because, you know, there were more people internally said, ah, it's not going to take that much time, let's just leave it up and running. But in general, it was kind of an afterthought at that point. 00:45:13 - Scot Wingo Yeah, there were some assets there. I remember it would. You had a lot of Twitter accounts for every sports team and the SEO was all through the roof because it was just like very SEO baity because it was so much content you could create websites for. 00:45:25 - Robbie Allen Yeah. One of the interesting things that we did was we created a website for every team. And so I was a big believer in team based coverage because that was my whole story. Right. I like college basketball. But really all I care about UNC basketball. I watched Kentucky beat Duke last night and I enjoyed that. But in general, I just care about unc. And so, like, I only wanted UNC coverage and I thought that's how a lot of people kind of think about sports. And so what we did was again, we were trying to automate a lot of this coverage and we thought, what if we just built websites for every single team? And so we literally had a thousand websites. Then we had Twitter handles, you know, the API. They had an API at that point that allowed you to post content automatically. And we thought, well, we'll do is just generate a Twitter sized version of a recap and do these things. And so we had Twitter handles, we had Facebook pages, we had websites, we had all these things. And we even started doing mobile apps or at least mobile websites for all the teams. And that was kind of a fun experience. We even created little logos for every single team, which got us into all sorts of copyright and trademark issues. 00:46:27 - Scot Wingo But sometimes doing something controversial is good. So you're into this narrative thing and then naturally you're in sports, but then you started to get a lot of inquiries into. I remember franchisees would come to you and say, or franchisors and say, you know, I'll make one up. I'm Subway. And every day I send this big data dump to 3,000 restaurants. And none of them read it. And we think that they would prefer a narrative kind of a model. So it's really interesting that kind of started with sports and then you, you ultimately did a cool sports deal we'll talk about. But you started to find other applications as well. 00:47:00 - Robbie Allen Yeah, and that was intentional. And one of the challenges, I mean, it's great to do something new. And like, for a lot of my career in starting companies, it's been a new thing, a new market. And that's exciting to get behind because not many people have done this thing before. The challenge is you're having to educate the market. And so we had to spend a lot of time, people didn't know this capability even existed that was possible. And then we'd go in and say, okay, here's some examples of. We took this data and when it generated this, and I remember like it was yesterday, the most common response was, you know, someone who read it and they say, okay, so who wrote that? I was like, well, our technology generated. It's like, I know, but this is like somebody wrote this article. I was like, no, no one wrote the article. And it's like, it Was we had this debate about, like, they didn't believe that technology could possibly do that. Right. And so that was a big hurdle. 00:47:49 - Scot Wingo Yeah. Yep. So then somewhere in there, you had a situation where one of your early investors kind of wound down or something. Right. They disbanded and you had to kind of go fix that. That was kind of a. A mess. 00:48:01 - Robbie Allen Yeah, yeah. We had one of our, you know, the. The firm that had done that original seed investment wound down and, you know, weren't supporting us in future rounds. And that was at the time, one of the arguments against doing large VC LED seed rounds was the signaling problem. And that was if you raise a large round or, you know, you raise money from a VC and then they don't come in for your Series A, then that's a signal to everybody else that, like, well, we didn't think enough of them to continue to follow on, and then maybe that could tank your ability to raise more money. That was generally, you know, why a lot of people wouldn't do those kind of deals. But again, for me, I was, you know, I wanted to bet on our ability to be successful. Now, again, that firm ended up winding down and they didn't support us, so we were going through that exact problem. But, you know, we. I won't say it was easy to get the Series A done. And again, we were doing all these changes. I look back on it now, I'm like, I don't even know how we were able to pull it off because again, we changed the company and the model, and then we didn't have our seed investor supporting us. And we, you know, the executive team had to go, I don't know, it's probably three or four months with no salary. We had to defer salary because we were just running up on the end of the money before we were raising the Series A. And then, you know, there was still the aftermath of the financial crisis. That happened, too. So anyway, it was a stressful time. 00:49:24 - Scot Wingo And then you. You ended up getting your A. Who. Who led that? 00:49:26 - Robbie Allen So Court Square Ventures, which, you know, they're based in or was based in Charlottesville, and so they. They were in that. And then we had a number of other VCs that came. Came in as well on that one. 00:49:40 - Scot Wingo Yeah. You remember, the early guys didn't know what to make of you. 00:49:43 - Robbie Allen Yeah. 00:49:45 - Scot Wingo They didn't like the consumer part of. 00:49:46 - Robbie Allen What you're doing, for sure. 00:49:47 - Scot Wingo Yeah. And they didn't understand the AI, the automated insights part at all. Yeah, yeah. No one, like, knew, like, they're like, okay, it felt like a toy to a lot of people because, like, what do you do with this? 00:49:58 - Robbie Allen Yeah, yeah. And, you know, we hadn't been in. 00:50:00 - Scot Wingo Those meetings where like, people were like, oh, my gosh, if I could send this to 3,000 of my franchisees, this would be way better than make forcing them to log into a dashboard they don't look at. 00:50:09 - Robbie Allen Yeah, that's right. And so it was a lot of hurdles. You know, again, there were only like a couple of us in the business doing that kind of thing back then. And so, you know, it wasn't like we were competing against Google or. And in general, there was less noise on the Internet, so we could get a lot of attention that way. But it was still a struggle to educate the market. 00:50:27 - Scot Wingo That was interesting. You had a company in Chicago called Narrative Science, if I remember, and they were. They were going at it. You were kind of going after different verticals and they were going after the newspaper industry. They were like going head on, basically saying, you know, use this instead of the AP or something like that. Yeah, yeah, whatever. We'll talk about that later. Okay, so you're chugging along and then you did your big deal with Yahoo. Yeah. 00:50:48 - Robbie Allen Yep. 00:50:48 - Scot Wingo How did that come to fruition? 00:50:49 - Robbie Allen So again, you know, it's these kind of lucky turns of event that you can never plan on. But we had talked to. We actually talked to all of the big sports companies, sports news companies, Yahoo, espn, the NFL, because we had this thought that, you know, where better to personalize sports coverage than fantasy football? You know, fantasy football was really big. Still is big. And, you know, people would devour any content they could create. Imagine you had your own personal recap of why your team beat my team and the great moves you made or whatever. That might be a way to kind of keep engagement high for these fantasy platforms that were all just duking it out, trying to, you know, get as many users as they could. So we went to Yahoo, and it was actually the GM of Yahoo Sports, David Geller, who I'm still friends with today. He took a chance and it was like, yeah, let's do this with Robby. And it was so funny because I still was good friends with the head of ESPN Digital, and I went to him and said, hey, we got this thing. We could do automated recaps for your fan. You know, tens of millions of fantasy people, they would love it. And he told his team, and ESPN was famously an nih, not invented here type of company, where they only wanted to build internally, they never wanted to use partners the head of ESPN Digital told his team, I want to do this with Robby. And we still never could get it done. And yet with Yahoo. David Geller, he was able to make it happen. And so we actually got a contract done through him. And one of the great things that we said is we wanted to have a byline in the footer. So at the very bottom, we just wanted to say powered by Automated Insights with a link back to our website. And so, sure enough, we got that done. And it was just this huge boon for the company because pretty much for 14 weeks of the year, we'd send out 10 million of these recaps. Well, guess who plays fantasy football. A lot of business people. And so they would see this recap, and the feedback was universally, they loved it. And at the very bottom, it said, powered by Automated Insights. You click on the link, go to our website. And so our traffic was just from September through the end of the year was always amazing. And it sent just so many people our way. 00:52:56 - Scot Wingo Yeah, I remember it was very stressful, though. So Monday Night Football would have to happen. And then after Monday Night Football, the machines would do their magic. So that by. And there was a pretty hard deadline where they wanted it all delivered by. 00:53:06 - Robbie Allen The next morning at, like, 9. Yes. 00:53:08 - Scot Wingo So you literally had, you know, less than 12 hours to generate 10 million unique narratives. 00:53:13 - Robbie Allen Yeah. And they had to send, you know, there were so many points of vote. They had to send us all this data. You know, they had to do all their own computations because they had to calculate points and things like that. They had to send us all the data. Then we had to generate all of the content and handle, again, there's 10 million of these recaps, all the different permutations and errors that could come out of different things. And so, yeah, that was super stressful. 00:53:33 - Scot Wingo Yeah. Many sleepless Monday nights, I mentioned. 00:53:35 - Robbie Allen Yes. And we called them work late nights where we have people come and, you know, we'd stay till 1, 2 in the morning. 00:53:41 - Scot Wingo Yeah. Yeah. Cool. And then. And that was a really. That actually had revenue with it too. 00:53:46 - Robbie Allen Which was very nice. 00:53:47 - Scot Wingo Yeah, it was a really big. They paid you quite constantly as well. That was awesome. 00:53:50 - Robbie Allen And then we at the same time did a similar deal with the NFL. So the NFL Network, we. We powered their fantasy football recaps, which is pretty cool. Got to go to the NFL headquarters and. 00:53:59 - Scot Wingo Yeah, so you're chugging along, you're doing all this. You'd done your Series A, and then you started to get some acquisition kind of nibbles. 00:54:07 - Robbie Allen Yep. Well, first we. I mean, we always kind of got different inquiries, you know, over the years, but then we raised our Series B, and that was in 2014, summer of 2014. And who led that one? That was Osage out of Philadelphia. 00:54:21 - Scot Wingo So at this point, you raised a million. What'd you do on the A? 00:54:25 - Robbie Allen I think that was four and then five. 00:54:28 - Scot Wingo Yep. 00:54:28 - Robbie Allen So 10 million, something like that? A little over 10. Yeah. And so what had happened just before we raised our Series B? And this is often why I tell entrepreneurs that you have to just be able to hang around long enough for the confluence of events to line up in your favor that would help you either get to the next level or get acquired or whatever. Well, what happened was that Summer Stats Inc. Got acquired by Vista, which is one of the large private equity firms, software acquirers. And Stats Inc. Was the, at the time, the number one sports data provider in the world. And so they were the big one. That was, you know, when I told you in 2007, I paid for my sports data, it was actually with Stats, Inc. Stats, Inc. So I had had a bit of a relationship with them over the years. And so they got acquired. Well, the private equity model is. Well, once they go and make a big acquisition like that, they want to then go out and find who else can they acquire to help prop up or improve the ability for that big acquisition to get to the next level. 00:55:25 - Scot Wingo They call it a platform. They buy the anchor platform, and they glom a lot of stuff onto it. 00:55:29 - Robbie Allen That's right. And so one of the reasons I like the digital sports space to begin with was there weren't many companies that were innovators in that space. In fact, there weren't a lot of companies. Now, it's partly because it's hard to monetize, but what that meant is there weren't other companies really out there to compete with. Cool new stuff. I was building at the time with Stat Sheet, and so competitor wasn't doing sports. 00:55:54 - Scot Wingo Narrative science was like doing, I don't know, business articles or something. I can't remember what they were doing. 00:56:00 - Robbie Allen And so when it got to be 2014 and Vista acquired Stats, it was pretty much still the same environment. That is there weren't a lot of innovative sports companies out there. And so when they. I can imagine the internal conversation they had, which was like, all right, now that we've acquired Stats, who else are we going to acquire? And there's like maybe three or four companies on the list, and one of those companies is, well, Automated Insights that, you know, they used to be stat sheet and they still do these fantasy football things. Maybe we should go talk to them. And so we had just completed our Series B that summer. This was 2014. And then by September, Vista, you know, contacted us and said, hey, we'd love to talk. And so, you know, I just raised our Series B. We had new investors that were super supportive. And I was like, ah, well, I mean, yeah, we can talk because I believe in keeping your options open, but it's nothing that we're all that interested in right now. 00:56:51 - Scot Wingo Conventionalism is when you raise around, there's this window of a year, you can't really sell the company because you've like, jacked, you know, the. That investor's expectation is a big return and you've got to kind of like execute past that to even get close to an exit. 00:57:03 - Robbie Allen That's right. And so I was like, I don't know if we should even look at this, but we thought, well, it's Vista and they're known for being pretty aggressive acquirers. And so we thought, well, at a minimum, we need to have the conversation. So I flew down to Austin and met with a bunch of the partners, including Robert Smith, the owner. And that meeting went extremely well. And I gave him the whole pitch on why the future is narrative and not dashboards. And, you know, how we can automate this kind of analysis and present it in language form versus just presenting numbers. And I saw Robert look over at one of his guys and he kind of gave him a nod and a wink and I was like, ooh, that looked like a good sign. And so we left and we went back and then sure enough, they wanted to give us an LOI and make an offer. And we were like, ah, we just raised a Series B. The bar is going to be pretty high for us to consider anything because we got new investors. And they said, well, let's. Let's see if we can figure it out. And. And so, you know, as they say, they made us an offer that even our new investors couldn't refuse. And the way they like to pitch it is like, look at their irr, right? Like, their IRR was amazing. Like, I still think that we had the best IRR for them out of any company they've had because they only had us for like six months. 00:58:17 - Scot Wingo Time component. Yeah, yeah, yeah. Cool. So you sold the company? 00:58:21 - Robbie Allen Yep. 00:58:22 - Scot Wingo Yeah. Have you ever disclosed, like a range of like, kind of what that was or. 00:58:25 - Robbie Allen Well, what got reported was 80 million. And so I won't confirm or Deny. 00:58:30 - Scot Wingo That this is one of those 100%, not like a sell 51%, roll a bunch forward kind of thing. 00:58:36 - Robbie Allen So I rolled some forward and so it was kind of a sweetheart deal, frankly. So they said, well, Robbie, you're kind of, you know, a. They did that. We were acquired through Stats because they needed the platform. Their model was you acquire. But what they told me is they want us to work with all their portfolio companies because again, all of their portfolio companies were data companies and they were trying to figure out what to do with their data and maybe Automated Insights could help them. 00:58:58 - Scot Wingo Almost seems like a PE firm would love to use this internally. Like, imagine getting instead of like this big book of like financial data, just like get a narrative for each company generated. 00:59:05 - Robbie Allen Yeah, we, you know, they said, well, how about this? We'll just, you, you continue to run the company and you know, you won't be a part of, like, we won't subsume you into stats. You just kind of run it independently as you have, and then we'll make introductions to all of our portfolio companies and we'll give you some more capital. And so I was like, wow, this is great. So yeah, I'll do that. 00:59:27 - Scot Wingo Did they have a vesting on your shares? Like you had to kind of stick around for two, three, four years? 00:59:33 - Robbie Allen They did, they did. And so it was like I rolled forward some percentage for a couple of years, but yeah, it was still a great result. 00:59:42 - Scot Wingo Before we go into the life after private equity. When you started Automated Insights, you were one of the few first time founders and even multi time founders that spent all this time on culture. Like, what, where did that come from and why did you do that? 00:59:55 - Robbie Allen It's a great question. And it was from my time hating the culture at Cisco. It was, you know, like if I looked at it and again, I learned that lesson about it's the, it's the journey, not the destination. And you know, I'd had a couple of experiences. I'd been pretty lucky in my career to work good people, but there was a couple times where, you know, I worked with somebody that was just completely, you know, one person could just completely zap the productivity and even just the overall happiness level of a team. I was like, I do not want that. Like, if I if part of the benefit of me going and starting a company is I can kind of mold it to be whatever I wanted. And what I want is to be an environment people enjoy coming to work. And so to me it was in many ways it was kind of a selfish thing as well as, like, make sense for the business for people to enjoy. I never kind of believe that people operating under fear would necessarily do better or, you know, especially if you think about it, sort of post 2010, it was really starting to become a very competitive environment for talent. You know, it was hard to keep people around. And so I thought, well, for us to be able to compete, we need. We're going to have some cool technology to work on, but we really need to have a culture that people enjoy working at because they're going to get offers for other places. And so why aren't they going to take those offers? Well, hopefully it's because they liked working with us as well. 01:01:07 - Scot Wingo Yeah. So I remember. So you had an office at American Tobacco, and it kind of overlooked the baseball field, which is cool being a sports company. You were like, right, you had those sports thing, but you walk in and there was like, there's definitely a ping pong table, maybe a foosball table, maybe a shuffle board. And then. But then, like, you also had on the walls, like, a bunch of vision statements like, like, how did you. What was the. What was the culture, like, statement that you had? And did you do, like, culture values? Like, how did you. I think this is. This is kind of interesting for folks to hear. 01:01:37 - Robbie Allen Yeah, it's mostly. Values were the focus, you know, so we had in, you know, over time. And one of the things I kind of learned was that it's. It's hard to grow the culture over time because you tend to want to add things to it, you know, not take away. And so, like, you know, we started off with like four values, and then it went to six and. And then it went to eight. And then I remember standing up in meetings and I would be like, okay, I want to review the values and I get through. And I'd be on like, number six or seven. I'll be like, and I'm forgetting the. What's the last one? I can't remember, you know, and it's like, all right, maybe there's too many. If you can't hold them in your head and repeat them that easily, then you have too many. But really, they were all just kind of statements around, like, what are we optimizing for? You know, what is it that, you know, automate everything? That was one of our. Our values. And it's like, you know, I wanted. I kind of thought about it. If somebody's sitting at their desk and they're like, all right, I got five things I could work on. What do I Pick, like, you should be able to run through the values. And it's like, okay, that gives me a good guidepost of how to prioritize what I'm working on as well as, you know. Another one was no jerks. You know, that was because, again, kind of getting to this point of I don't care if you're, you know, I always said, I don't care if you're a rock star engineer. I wouldn't take you, you know, if you're a jerk, because we don't need that. Right. That's not going to. You're going to end up hurting more than you're helping. And I don't think we're going to be successful because we got the one person that just knows how to, you know, that can work 80 hours a week. You know, for us to be sustainable, we need a whole team effort around this. And so it was really, you know, these kind of different values focused on different parts of the culture and what we were trying to optimize for when we were hiring people. 01:03:11 - Scot Wingo Yeah. And I remember you had, like, these epic ping pong tournaments that. And there was like a whole thing there that the whole team was very compatible with each other, but in a fun kind of way. 01:03:19 - Robbie Allen Yeah. And I believed it was good to have an environment that people, again, kind of enjoyed. It wasn't just about work. It was also getting up and moving around a little bit. And we might have overdone it. Automated insights. Because there was points in which I was like, I don't know, they seem to be playing a lot of ping pong, but we had people that would stay really late, too. And it's like, all right, if they're working, if they're here 12 hours and they play ping pong for two hours, we're still coming out ahead, you know. But it was. I mean, it got to be pretty bad. I mean, you know, one of my really good friends, Adam Smith, we were. This was in one of our first offices, we had a foosball table. And foosball is really loud. Right. That's one of the things about it. But we had it in our main conference room and we had. We were interviewing like an engineer. And so the. The people were interviewing him in that conference room. And Adam and I were like, well, let's play foosball. If they can't handle this while they're being interviewed, then this isn't going to be the right place for them. And so we were like, playing. It was loud and they're trying to talk during the interview, I was like, I don't know what we were thinking, but it's kind of funny to think back on that. 01:04:20 - Scot Wingo Yeah. Yeah. Another interesting thing about Automated Insights is a lot of people have gone to be entrepreneurs themselves. So that's. That's one of the great hallmarks of an entrepreneur is if you kind of get other people into the. Into the world. So Adam has wrangle now. And then his other Adam, he was in Automated Insights. So two Adams. Yeah. Adam Long, and then who else? 01:04:38 - Robbie Allen He's got something called Hemingway, which is a writing app. 01:04:42 - Scot Wingo Yeah. And then any other folks. 01:04:44 - Robbie Allen So there's a bunch that went on to other companies that have also marketing. 01:04:49 - Scot Wingo Folks and your finance lady. That may have been. Always get confused with Windsor Circle. That may have been. But yeah. So, yeah, you had a lot of really good folks that have stayed in the community. 01:04:59 - Robbie Allen Yeah. 01:04:59 - Scot Wingo For sure things. Yeah. Okay, so Vista's acquired you. Now you're working for the man. How did that go? 01:05:04 - Robbie Allen You know, it's funny because I always remember hearing about, you know, people that would stay one to two years before, you know, post acquisition, and they'd leave. And when Vista made me that deal about, like, you just keep running the company the way you want to and we'll give you more money to run the company, I was like, that sounds perfect. Why wouldn't I just keep doing this? But sure enough, after two years, I was like, I don't know, I think I'm ready to do something else. And so I don't. There's something about that, you know, whether it's loss of ownership or, you know, you're not exactly in control in the way that you were before. 01:05:34 - Scot Wingo Yeah. 01:05:35 - Robbie Allen That after a couple of years, I was like, you know what? Maybe. And for me, you know, I started stat sheet in 2007, and then it got to be 2017 or 2016, and I was like, ah, it's a pretty good run. 01:05:47 - Scot Wingo Two years in. Yeah. 01:05:47 - Robbie Allen Yeah. I'm not. You know, another big thing that I coach, you know, especially early career people on, is I don't actually. I don't believe in staying somewhere 30 years. I think that's the worst thing you can do. I think it's okay if you're just cool coasting and like, again, you don't really, you know, you're not too worried about the consequences that we talked about earlier where, you know, you're losing skills and network and all of this. And maybe one day you get laid off after 20 years and you're kind of left holding the bag. I believe it's good for people to change up every five, six, seven years. And that's in across a bunch of different aspects. I mean, especially in your career, it's good to change things up pretty dramatically every five or six years. And so I'd been, you know, running stat sheet and automated insights for, you know, going on 10 years. I was like, that's a good run. I'm probably. It's probably time for me to try something else. 01:06:34 - Scot Wingo Yeah, cool. So you come out of there and you decide, you go back to your old playbook and what do you do? 01:06:41 - Robbie Allen Yep. So I was like, well, what am I going to do in my sort of mini retirement? You know, at that point I was, I don't know, 40s, early 40s. And I was like, well, I can't just retire, you know, I'm got to be too active for that. And. But I'm not ready to start a company yet. And machine learning has really taken off, you know, I. That's something I'd studied, you know, a long time ago. And, you know, there had started to become more sort of interesting developments there, you know, just in terms of how neural nets had progressed and machine learning models are really starting to show some improvement. And I thought, well, it would be cool is if I went and worked on my PhD in machine learning. And I know, how about I do it at UNC so I can finally answer that question. I get at dinner parties all the time, which is, oh, what year did you go to unc? And I had to tell the whole story again about, ah, I got in, but I didn't go and all that. So I applied and they accepted me for the computer science PhD program at UNC. And I was, you know, living it up. You know, I'm on Franklin street getting coffee, going to classes, and of course I'm the old dude in the class, you know, but I was really enjoying it. 01:07:46 - Scot Wingo Yeah. So you decided to get a PhD at UNC and then what happened? 01:07:52 - Robbie Allen Yep. So pretty shortly into that, the head of research at Duke University approached me about starting a machine learning company. And. And again, I was like, I was a pretty hard sell on it because, you know, I was doing what I wanted to do. 01:08:06 - Scot Wingo How did you know, so you're at unc, but the Duke guy, like, how did that work? 01:08:09 - Robbie Allen Well, so, you know, they were wanting. So what had happened was Dr. Larry Karen had a large lab of PhD data scientists at Duke, and turns out Larry was one of the 10 most published researchers in the world on machine learning. Him and his lab. And so again, this machine learning was really starting to take off, specifically deep learning, which was kind of a relatively new form of machine learning, where you apply more complex kind of parameters and data to these models. And they could automate more stuff. They could automate more complex processes. 01:08:41 - Scot Wingo But those are neural nets on the. So they're, they're, they're inferring kind of the connections between whatever the data is and stuff. Yeah, so they're learning. 01:08:48 - Robbie Allen Yeah, yep. 01:08:49 - Scot Wingo And then this was so at the time, a big seminal moment in the AI world was you had the whole Watson beating jeopardy. But the more interesting One was when DeepMind, the Google DeepMind people, when they want it go. That was like a big, I always hear AI people talk about that. Was that the same kind of a learning models kind of thing? It was. 01:09:10 - Robbie Allen I mean, that happened a little bit later than when we started. I think that was 2017 or 18. But yeah, it was the same kind of thing and that. And there had been a lot of kind of developments before that one called Alexnet, you know, these other things where it could predict what a picture was. You know, there had been a lot of advancements happening in that space. And so there was kind of two different areas. There's image based processing and then text processing. And both of those are kind of happening in parallel, but both showing kind of lots of interesting results. 01:09:43 - Scot Wingo And so this Duke, Duke calls you. 01:09:46 - Robbie Allen Yep. And so he had this large lab of data Scientists and data PhD data scientists, especially if they were publishing at these high end machine learning conferences, were getting a million bucks a year from Google and Facebook and some of these places. And so they were very prized in the industry. And at Duke they had a whole lab full of them. And so the idea was, why don't we take this expertise and go apply it to the business world? Because we don't know, like, there's this new capability now where we can automate processes that previously couldn't have been automated because we have these complex machine learning models and we can throw. Now that big data has been a thing for a lot of years. Companies have enough data that we could train the models on and maybe we could automate very valuable processes. But going into it, you know, again, they're, they were an academic lab. They, they didn't know which ones to start with. And so the idea was, well, why don't we just go out almost like a consulting company and do a bunch of projects and then after a year or two, we could kind of look around and see which One stuck or which ones do we think were the most valuable and go productize that. 01:10:48 - Scot Wingo Yeah, and these kind of deep learning models do better if there's a known like a, an in, like an end result, like you know, bidding it at, you know, bidding on ads and you know, you want to pay this and you got a sale. So like an end point. So they can kind of like feed that back into the model. 01:11:03 - Robbie Allen Right. 01:11:03 - Scot Wingo So so a zero sum game of winning and then like, you know, understanding the implication of that. 01:11:08 - Robbie Allen Yeah, yeah, it's. Especially back then it was more of, you know, the training data, which is, you know, given this input, this is the output that you want versus something like, you know, autonomous driving, which is like, it's, it's almost more free form and. Yeah, and you have to figure it out as you go. Yeah. And so they wanted to go start this company, but they needed a CEO to run it. And there was just a short list of local CEOs that had done anything related to AI. And I was on that list and so that's why they approached me. And so, yeah, it was something I wasn't exactly sure of initially because again, I was having a really good time at unc, but this was, I kind of viewed it as a once in a lifetime opportunity just to have the kind of quality team that we'd have. We also got backing from a private equity firm. So a private equity firm was going to put $10 million in if we could get this all situated and so we'd have the capital. 01:12:01 - Scot Wingo We just do this like one of the most published people, one of the. 01:12:04 - Robbie Allen Most at the time when it's probably peak value to have somebody like that. And so I was like, maybe this is worth giving a shot. Totally different. Again, my previous company, I was the technical solo founder. Right. And so I was used to developing the technology and running the company. And this was. I'm kind of coming in after they had already thought about starting the company and I wouldn't like, in fact, they probably wouldn't want me coding because like, compared to these other PhD, like, I don't have my PhD yet. Right. They already have theirs. 01:12:33 - Scot Wingo Yeah. Did they have like an engine already or. There was essentially, it was just like more conceptual. Here's how you apply this technology to a problem. 01:12:41 - Robbie Allen And so that's, that's how Infinea ML got started. 01:12:45 - Scot Wingo Cool. What year? 01:12:45 - Robbie Allen So this was 20. 2017. 01:12:47 - Scot Wingo 2017. Okay, cool. So then give us the quick summary of how that went. 01:12:52 - Robbie Allen Yep. So, you know, long story short, machine learning doesn't Generalize. Well. And what that means is we did a bunch of interesting projects. You know, we. In fact, we got one done with Walmart, which I kind of view as sort of a resume item, just because it was so difficult to get projects done with Walmart. They're notoriously difficult. 01:13:10 - Scot Wingo But go to Bentonville and see that situation. 01:13:12 - Robbie Allen Yeah, yeah. Which is pretty crazy. 01:13:15 - Scot Wingo Did you return your pencil at the end of the meeting? 01:13:17 - Robbie Allen Yes. So we did a bunch of projects, some with the Department of Defense, you know, just all different kinds of projects. Kroger, you know, different things. And after a couple of years, you know, we were having a board meeting, and the, you know, private equity firm that was on the board along with me was like, okay, so a couple years in, you know, money's gonna run out. I mean, these projects we were doing, pretty profitable. But, like, we had a lot of, you know, 15 PhD data scientists that were very expensive. So the burn was not trivial, Right? And so we were burning, and there was going to be an end to the cash. And the question was, well, what do we. You know, the whole reason to bring me on was to build a product, right? Like, I'm not a consulting CEO. I'm a product CEO. And so after a couple years, we sat down and said, all right, well, what product are we going to build? And, you know, I had to kind of bring the harsh reality that, well, we've done a lot of interesting things, but nothing kind of is rising to the top that we should go build and focus all of our efforts on that. And they're like, huh, well, that's not great. And I was like, well, you know, this is where we're at. And they said, well, how much money will you need and what can you promise in a year from now? And I was like, well, as Jeff Bezos says, we are in the wilderness, and I can't really promise a whole lot of anything in a year. I mean, we're methodically going through all these different applications of it. But unfortunately, deep learning is highly sensitive to data variability. And what that means is you could have the exact same application for automating a process in two companies, but if the data is off just a little bit, then it kind of invalidates the training that the original model was based on. And so what you need is pretty much this exact same kind of data. You know, the example I like to use is facial recognition that you can train one model, you can go apply it anywhere. And if you trained all on faces, you know, it doesn't matter. You can Apply that in lots of different ways. But if you're talking even sales data, unfortunately sales data can vary and does vary between every company, just the semantics of it, how they use it, you know, what the leak. 01:15:20 - Scot Wingo So you couldn't take the Walmart thing and apply it to Target because there's enough differences that it had to be bespoke for every. You had to go retrain the whole model and basically wipe, you know, restart kind of thing. 01:15:29 - Robbie Allen That's exactly it. And so, you know, after, you know that, to me it was pretty clear it was going to be a while before we could figure that out. That that wasn't kind of what I signed up for. And so I said it's probably best to find somebody that could kind of run this more as a consulting operation until you really find a product. And so I said I'll stay on as an advisor and continue to help the company, but I'll step back. And so we agreed to that. It was lucky timing for me because this was winter of 2019 and then Covid happened like four months later and pretty much everything shut down. But at that point I was no longer the CEO and so I didn't have to run a company during the pandemic, which was again good for me, but hard for a lot of other people. 01:16:13 - Scot Wingo Yeah. Were these machine learning things pretty compute bound? Did you have to like have a, your own bank of GPUs or could you run it on AWS? 01:16:20 - Robbie Allen We, we did need. And we had our own local GPUs, but we also use cloud as well. But that was certainly a factor. 01:16:28 - Scot Wingo Yeah, interesting. 01:16:30 - Robbie Allen But you know, we weren't like, we. 01:16:31 - Scot Wingo Weren'T bought Nvidia stuff. 01:16:33 - Robbie Allen Absolutely. But we were also weren't training on like Internet sized data. Right. Like a lot of people make a lot of, you know, how much compute's needed. That's because they're training on so much data. You know, you could do you, you know, at the time we only needed a few Nvidia machines and you could actually train a reasonable model on quite a bit of data. It's just when you're talking Internet scale, that's a whole nother ballgame. 01:16:53 - Scot Wingo Yeah. 01:16:54 - Robbie Allen Cool. 01:16:55 - Scot Wingo And then ultimately they got acquired at some point. 01:16:58 - Robbie Allen Yeah, Esperian, you know, acquired them a year and a half, two years ago. 01:17:03 - Scot Wingo Cool. And is the professor still there or is he back at doing this thing? 01:17:07 - Robbie Allen He's back at Duke now. 01:17:09 - Scot Wingo Cool. 01:17:10 - Robbie Allen From last, last I heard. 01:17:11 - Scot Wingo Yeah. So then you're. We're in the pandemic and then you decide That's a great time to start a health company. 01:17:17 - Robbie Allen Yep. Well, I first, me and my wife homeschooled our kids during the pandemic and traveled. So we, we're like, you know, we always wanted to travel Europe for extended period of time. You know, my daughter's 17 now and my son's 10. So the time was kind of running out that we could make that happen. And I was always, always in a startup or a company. And so we thought, well, if we're ever going to do it, now's the time. And so we homeschooled them and traveled Europe for several months and had an amazing time. But then coming out of that, I was like, all right, now I'm ready to start a company. I need to get back to doing something work related. And so that's when I started thinking through what I wanted to do next. And this time I wanted it to be a little bit different. You know, I'd gotten into my own personal health quite a bit over the last several years. You know, just focused on, you know, trying to. To stave off some of the things that, you know, I was seeing around me, whether it's friends and their parents having issues or, you know, just somebody like one of my daughter's. My daughter's best friend in, like, fourth grade, her dad. It was a Sunday night. I remember, like, it was yesterday. So she was in fourth grade. And what they. They said was the friend's dad was walking around upstairs, and they were downstairs kind of getting ready for, you know, the coming school week. And all of a sudden they heard a loud thump. And they're like, what's going on with that? They went upstairs and the dad was dead. He's like 40 years old. 01:18:37 - Scot Wingo Holy cow. 01:18:38 - Robbie Allen And just died of a heart attack. And it's like, wow, that is crazy. And obviously, you know, she had two small kids and now she's got to, you know, go without, you know, the dad being around. And so that was traumatic. And, you know, just seeing things like that, I was like, there's got to be a better way to tackle this than what we, you know, my yearly physical where my doctor really doesn't do much of anything. And yet I'd been following people like Peter Attia and Andrew Huberman, and sure enough, there's a lot of new research and studies coming out showing there's actually things you could be doing now to help so that you could be healthier for longer. The problem was there's not many doctors around that can help you on that journey. 01:19:15 - Scot Wingo And so, and like anything on the Internet, there's always conflicting data. 01:19:18 - Robbie Allen There's conflicting data. You know, you really needed somebody that's, you know, a physician that's trained in that to help you understand what to do with it. But then also there's all this data and I figured, you know, there's a lot that machine learning could probably do to help you understand that data. And so those kind of came together and I approached my co founder, Dr. Jared Pello, about starting a new preventative health company. And I couldn't do this on my own. You know, I considered, I think being a solo founder is the ideal situation if you can make it happen, but it requires, you know, you understanding the market really well. You know, you kind of have to be able to check all the boxes, you know, for the company. And that's not always easy to do. And in this one specifically, like it's a health company where we were going to actually be providing care to people. I'm not a, you know, a medical doctor and so I needed somebody that had that understanding. Well, fortunately I did. And my good friend Jared, who I had actually worked with, you know, several years before when he had his first startup called iScribes and I was his independent board member and I kind of, you know, oversaw that as they were acquired by Nuance Communications and then that got acquired by Microsoft. And so Jared had been at Microsoft the last couple of years and so I approached him and we met at the Starbucks off of Slater Road. And I said, hey, why don't we go start a company together? I hate to see you waste away inside of Microsoft for the rest of your career. 01:20:40 - Scot Wingo Yeah. And I scribes was interesting because it was natural language, like, you know, that's what Nuance had. Dragon something. Yeah, Dragon, yeah. So that was like a. Interesting. So he kind of like auto. He kind of already was primed to talk about AI stuff in an interesting way. 01:20:54 - Robbie Allen Yeah. And that's why he asked me to join his board back then was just because that I had that experience. 01:20:58 - Scot Wingo Pretty cool. Did they raise money or was that Bootstrap? 01:21:01 - Robbie Allen They had raised money, Yeah, a little bit. 01:21:02 - Scot Wingo Yeah. It's funny, they kind of like were off on their own. Like a lot of people don't remember them as part of the triangle startup scene. So you start Bionic Health and then what? You know, for folks like what. What's the idea? 01:21:15 - Robbie Allen So the idea is we provide proactive healthcare for busy professionals. And so what does that mean again? You know, fortunately I don't have to, you know, this one doesn't require a lot of education like my other companies did. It's, you know, I just say, do you enjoy the yearly interaction you get with your primary care doctor? And almost, you know, everybody I come in contact with says yeah, I mean, well, we don't really do much. I see them for 10 minutes, they do a little bit of testing. And I said, well, there's actually a lot more that you could be doing now. And if you've ever had again a parent go through something and they just aren't in good health when they're in their 70s or things like that, odds are there's things they could have done in their 40s and 50s so that they were healthier then. And we know those things now. They didn't have the benefit of that knowledge back then, but we know that now. And so what we do is provide a physician guided process to help do the testing and help you put together a treatment plan so that we can extend your health span, which is the number of healthy years that you have. 01:22:12 - Scot Wingo And then is that a thing that you guys came up with or that's like a known, kind of like the Peter Attia guy or how we say it, he did. Or like how is that, how is your healthy lifespan kind of determined. 01:22:25 - Robbie Allen Yep. So it's, you know, it's kind of been growing over the last, I would say 10 years. You know, again it's been a little bit on the fringe, but now it's become way more mainstream. There's a book that Atiya wrote, wrote, called Outlive that's super popular and if you've read that, then you totally understand what we do. That's essentially the embodiment of what we provide at Bionic Health. And now they're starting to become other companies that are doing that. One of the nice things about the healthcare space is it's not a winner take all market. So just because somebody else tries to offer a similar service in another location doesn't mean it necessarily takes away, away from our ability to do that because there's so many people that could benefit from the application. But yeah, it's still kind of, it is early days. So it's not like we have to educate the market in that some people are still stuck around the idea that, well, I get my annual physical for my primary care doctor, isn't that enough? And it's like, well no, there's actually, that's not enough and there's quite a bit more that you could be doing. And it really just depends on how much you want to try to extend those healthy years. 01:23:25 - Scot Wingo Another thing that kind of stinks about our healthcare system is it's all geared towards when things go wrong and nothing preventative. Right. So if people want to do this preventative stuff, it's usually out of pocket, unfortunately, but usually you can do it without, you know, most of its lifestyle changes. Yeah. So what are. So for listeners that are interested in this, what are, like, four top things people can do to extend their healthy life? 01:23:47 - Robbie Allen So there's three things that, you know, three options that you can look at to. To help you improve some aspect of your biology. First, there's lifestyle change. And so that's eating healthier, working out, sleeping better. Obviously, those are the kind of fundamentals. There's supplementation, and so, you know, helping kind of supplement areas where you're deficient in certain nutrients or vitamins. You know, that can certainly help. And then the last is medication. And then, you know, we've seen everything from GLP1s to hormone replacements to, um, you know, other things that help reduce cholesterol and chances of cardiovascular disease significantly. Um, you know, there's a whole bunch of things like that that's available now, kind of depending on where you're at and what aspect that you want to try to improve. 01:24:34 - Scot Wingo Yeah. So to get started, people go get some blood work, and then you guys analyze it and you kind of say, here's your roadmap for. For improving this. 01:24:42 - Robbie Allen That's right. And. And we just start with blood work. And that's where most physicians kind of start and stop today. But there's a lot of additional testing. And, you know, as you know, with. With our experience with the Tweener Fund, I'm kind of, you know, any sort of diagnostic that can help collect data on your biology, I'm a sucker for. Because, you know, it's not just blood. I mean, obviously, blood's probably the most, you know, informative and the easiest one. But then there's also things like DEXA scans, which is a full body composition scan that tells you your lean mass versus fat mass. There's things like continuous glucose monitors that analyze your glucose over time based on what you eat and how that impacts things. So there's lots of devices now that help you understand your biology and what you do and how it impacts that. 01:25:28 - Scot Wingo Yeah, one thing that they. So on the all in podcast, they talk about, there's a company that will do this full body scan that's really expensive. It's like thousands of dollars. But you can go get this Relatively inexpensive calcium skin, and then it looks at your arteries. So tell folks about this one. 01:25:43 - Robbie Allen Yep. So there's, you know, there's kind of this order. If you go and say, you know, you've only ever done the typical annual physical, maybe your cholesterol is a little bit high, but you don't know, like, how bad it is. You go and get a calcium score, which kind of gives you a rough guide of if it comes back non zero, that means you have some sort of buildup going on. Um, and if it's non zero, then you need to get a cta, a CT angiogram, which is actually something that looks more specifically at your main arteries to see is there actual blockage or not. And those are all relatively inexpensive and pretty straightforward to get. Now, again, the part of the problem is you go into your primary care doctor and you say, hey, I want to go get a cta. They're going like, you're fine. You don't need that. 01:26:25 - Scot Wingo Yeah. 01:26:25 - Robbie Allen Or you're like, I want to get a DEXA scan. Why do you want to get a dexa? 01:26:27 - Scot Wingo That was my experience, too. 01:26:28 - Robbie Allen And so it's, you know, it's like you said, it's all about sick care. It's all about reactive. It's nothing about proactive. They're not trying to help you look around the corner. Our healthcare industry is great if you get sick or you have something that hurts, but if you're. If you're okay otherwise, then they're not going to help you. A great. One of the best examples I have is actually came from me. So when we first started doing this testing for bionic health, and obviously, I'm customer number one. Right. The reason I wanted to start the company is because I'm into this stuff. And so anything that we do, I want to try it out. So we did the full blood run up on me, and it came back to my ferritin levels were high. I was like, ferritin, what is that? I don't even know what that is. You know, good levels are below 200, and mine were over 900. And so that's not good. And. And so after digging in on it, what it. What it means is ferritin is your body's predisposition to store iron. And so what happens is some people store iron, which is a necessary thing that your body needs. It's a part of your blood, but you don't want to store too much of it because, you know, too much iron is really bad for your organization, organs. Well, what Happens is if you have a condition where you're storing too much iron over time, you won't manifest any symptoms for years and years and years. But then you, you know, turn 60 one day, and all of a sudden you start having organ failure. And then they discover, oh, you have too much iron in your blood, and we now we need to chelate it, which means extract the iron out of your organs. And that. 01:27:46 - Scot Wingo That's sound fun. 01:27:47 - Robbie Allen That's as pleasant as it sounds. 01:27:49 - Scot Wingo That sounds like Wolverine getting his skeleton. 01:27:52 - Robbie Allen Yeah, probably looks something like that. But then, you know, at that point, it's. It's, you know, gonna be very difficult. And so this is something that with a $10 blood test, you can find out if you have a predisposition to that when you're in your 40s and then start doing something about it, or you can wait until it's too late. Right. And so, sure enough, I. I found. I had that. I started donating blood. That's kind of the only way to. To lower your ferritin. And I just actually had my. My last blood check a couple weeks ago, and my levels are down under a hundred now. So I went from warning zone to totally safe. You've been donating blood like crazy nine months. You can. You can only do it every eight weeks. Yeah, but I've been doing it every eight weeks for the last nine months. 01:28:31 - Scot Wingo Wow. 01:28:32 - Robbie Allen And. And by the way, donating blood is actually really good for a lot of reasons. It's good for your body, it's good for other people. But, you know, that got me down, and now I'm in the safe areas. And again, that was something. I'm just walking around with it. I don't. You. You don't. You're not manifesting symptoms. Right. Your primary care doctor is not going to think, I want to test that just to make sure. 01:28:48 - Scot Wingo Yeah. 01:28:48 - Robbie Allen But again, that's one of the problems. 01:28:50 - Scot Wingo Interesting. And the other thing you do that's kind of interesting is you publish all your personal health data out there. It's kind of. A lot of people would think that's kind of weird, but. 01:28:57 - Robbie Allen And they do. And I don't exactly know why, to be honest with you. For me. So, you know, if we have success stories, one of the challenges in healthcare is like getting people to talk about, oh, yeah, my hormone levels were at 100 when they should have been at 500. And now we got them up. And so I was thinking, well, I don't really care if people know about my data, and so why don't I just publish all of it. And so if I have, you know, anything I could talk about, I can show it. So if you go to bionic.robbieallen.com It's a Google sheet with all of my data and it's all my wearable data, my blood data, body composition, you name it. I don't know of anybody that has more data publicly than that. Even the Brian Johnson's of the world and all of that. I doubt they have as much data publicly available as I do. But you know, I don't see the downside to it. 01:29:41 - Scot Wingo You've taken extreme transparency to a whole new level. 01:29:43 - Robbie Allen That's right. 01:29:43 - Scot Wingo Yeah. Do you, do you do also do like the 23andMe stuff to look for markers and genetic side. 01:29:48 - Robbie Allen I have. So I've done whole body genomes, not through 23andMe, but we're one of those. So with, with Bionic, we actually offer, you know, you can do whole body, I mean whole body, you can do whole genome sequencing. Again, all of those different areas where, you know, whether it's, you know, looking at your blood pressure, looking at your body composition, looking at your blood, looking at your DNA, you know, we want to offer something so that people can do that under the guidance of a doctor. So. Yeah. 01:30:16 - Scot Wingo Yeah. Cool. And then the other thing that's interesting about bionic from a software perspective is you have a clinic. 01:30:21 - Robbie Allen Yep. 01:30:22 - Scot Wingo Yeah. Is that, do you get to wear a white coat and have a stethoscope? 01:30:25 - Robbie Allen Only at Halloween. So last two years I've dressed up as a doctor and you imagine people just love that. They love me coming in as a doctor, but I'm Dr. Allen on that day. But otherwise I, I don't play a doctor. 01:30:36 - Scot Wingo Yeah. And then the, that's over in Chapel Hill. 01:30:39 - Robbie Allen Yeah. So we don't have like an in person clinic per se. Everything's done virtually. 01:30:43 - Scot Wingo Okay. 01:30:43 - Robbie Allen Right. So we do everything virtually. Um, you know, there is one thing that we offer. We have a performance module that we offer where you can get a VO2 max test, you get strength testing and kind of plot that out on a longevity curve to see where you're at if you're on track to live to when you want to be. Um, but everything else is done virtually. 01:31:01 - Scot Wingo Yeah. What's a VO2 max and why should people care about it? 01:31:04 - Robbie Allen Yeah, it's your ability to process oxygen under high loads and it's one of the best markers to assess, you know, how long you can live. Um, and it, it drops precipitously over time. Uh, you know, Obviously, I'm a big proponent of testing frequently, um, and frequently can be different for different people. But you know, one of the things that, that we found, especially with the mush testing as I'm doing, is people think you go and do your annual physical and I get the blood done and like my data point is at this level. And you think, okay, and I'll come in next year and my blood is at this level. And so it must just be a straight line from there to there, when actually it kind of depends on the biomarker. Some of them are more stable than others, but others, especially when it comes to hormones like it, they're bouncing all over the place. And so it's actually a fallacy to think that they're just connecting the dots or the actual trend of how your biomarkers are performing. You need to do more frequent testing to see throughout the year, you know, how things are trending. 01:31:59 - Scot Wingo Yeah. 01:32:00 - Robbie Allen Cool. 01:32:01 - Scot Wingo So you founded three companies. You're obviously, you know, I started the Tweener Fund and I got in over my head and I was like, you had kind of said if you can ever help, you let me know. I was like, I need help. Because it ended up being bigger than I thought it would be and we had to invest a lot more companies. So it's been fun working with you on that as well. Any what's your top kind of advice points that you give other founders? When the other thing that I appreciate is you spend a lot of time with first time founders and you're always paying it back, which is good when people come to you. And what are some of the top learnings that you share? 01:32:36 - Robbie Allen Yeah, I appreciate having the opportunity to work with you on this. Sometimes I'd get questions, especially from investors that I'm talking to about Bionic. They're like, oh, but what you're doing with this Tweener Fund thing? And my answer is always the same. My side hobby is just to meet with entrepreneurs anyway. Like, I'm gonna, I've met with, you know, I meet with entrepreneurs almost since the day I met with you. Right. Like, I think you set a great example back then about meeting with this guy out of a big company to talk to him about something he may want to start. And, and I really get a lot of energy from it, talking to entrepreneurs about their, their idea and, you know, hearing the passion. And so regardless of whether I was doing Tweener Fund or not, I was going to be meeting with entrepreneurs. Right. And so when we started talking about the Tweener Fund. It made a lot of sense for me because, you know, I was doing angel investing, but not very well. You know, I was just kind of scattershot. You know, I was meeting entrepreneurs, but maybe not feeling like I could have the impact. And the Tweener fund afforded me the opportunity to really focus my efforts around this movement you were creating around, you know, let's create something triangle focused. That can not just be providing advice, which of course we do, but then also even capital invisibility. I mean, that's increasingly something that I hear a lot of the portfolio companies enjoy is that when we talk about who we're investing in at the end of the quarter, they get a little bit of lift from that. And so in general, that's just been great and it's been a great outlet for me to be able to do that. In terms of the main advice I give entrepreneurs, it tends to be the people that very early on, just like with me, I don't really believe in regrets, but if there's anything I would have done differently in my career is probably started a company earlier, I just didn't think it was something that was accessible to me or something I could do when now what I think is anybody could be an entrepreneur. I don't really believe it's a DNA thing or there's some people. I think you can learn the skills. It depends on the type of company. I wouldn't say everybody's meant to go build a billion dollar startup company that requires a very specific mindset and skill, but in terms of starting a company, I think virtually anybody could do that. And it's really just a mindset thing. And so it's just a matter of do you have the ambition to do that? You know, do you have the will and the, as Hubert Davis says, the coach of the UNC basketball team, the will and the want to. And if you have those, I think anybody can be an entrepreneur. And so what I spend most of my time doing is alleviating the concerns people have. Well, what if I'm going to leave my nice cushy job? And I said, well, it's maybe cushy now, but maybe you're too safe and kind of working through all of those things because I think it's a viable opportunity for just about anybody to the point that even with my kids, what I told them pretty early on was like, I'm not going to have a college fund for you. I'm going to have an after school fund. And you can use that money in one of two ways. You can either go to whatever college you want to and I'll pay for your undergraduate, or you can go start a company and I'll be your first investor and I'll help you start the company. Now, my daughter, she's not going to take me up on that, but I still my son. Maybe. We'll see how that goes. 01:35:40 - Scot Wingo But I think a little Peter Thiel experiment. 01:35:42 - Robbie Allen That's Right. 01:35:44 - Scot Wingo N equals 2. 01:35:45 - Robbie Allen Yes. But I think even at 18, you could go start a company and get. College will always be there. You could go, you know, go to college later. I don't think it has to be then, you know, and again, it kind of takes the right personality and all that. I'm not saying everybody should do it. I'm saying everybody could do it. 01:35:59 - Scot Wingo Yeah. 01:36:00 - Robbie Allen If they wanted to. Cool. 01:36:02 - Scot Wingo Well, thanks for being so generous with your time. I know you're busy building bionic. Good luck with that. It's gonna be exciting to see the outcome there. And thanks for joining us today. 01:36:09 - Robbie Allen Yep. Thanks, Scott. For more Tweener content, check out the Triangle Tweener time substack@tweener.substack.com com. For more tweener content, check out tweenertimes. Com. Thanks for listening and we'll see you again soon on Triangle Tweenertalks.