Subscribe
Share
Share
Embed
AI After Dark is a podcast hosted by Alex Gras, venture capitalist at Mercury, focused on how real companies are built once the hype fades and the hard decisions begin. Through candid conversations with founders, CTOs, and operators, the show cuts through buzzwords to talk about people, systems, risk, and the trade-offs that actually matter. Alex brings a background as an operator, founder, and revenue leader, with a belief that technology matters, but people come first. This podcast is for builders who care less about trends and more about what lasts.
00;00;00;18 - 00;00;24;11
Unknown
We're company was this studio. So, come to school studio. So it started out as, a food company, which was a dumb idea. And so we still called studio, but food company. Well, the the company name is Morsel, Inc. So morsel was the food company we were trying to compete with the likes of. I don't know, like Maple and a variety of other kind of food companies that were hot in the middle there.
00;00;24;11 - 00;00;40;05
Unknown
2015, 2016, 2017. And we realized it was instead of 50% tech and 50% food, it was 99% food and 1% that if we're being generous. So we cut my wife. It's one point said, so you're running a catering company? I'm like, this is not what I want to do. So we shut that down. But it's fun.
00;00;40;05 - 00;01;02;13
Unknown
Yeah. And we pivoted. And my co-founder, Jason at the time was like, we were excited about the idea of charging for content. And, you know, we said, what kind of content might people pay for? And, you know, this fitness content was kind of an example of that. So we leaned into that initially focusing on audio running classes, saying, let's be really, really narrow.
00;01;02;15 - 00;01;31;05
Unknown
And that's where we started. And their first recordings were iffy and, we crossed copyright lines we didn't realize were there, which kind of got some trouble for period with Sony. Eventually we kind of figured out how to, you know, write all those wrongs and kind of getting the right lane and do things properly. And over time, we moved from audio running to video across six modalities or maybe more, and have a professional studio and record 25 000 classes, and you were selling to consumers as B2C.
00;01;31;07 - 00;01;51;25
Unknown
And so we were as B2C. So people would pay a subscription either $15 a month or hundred dollars a year. And then we would attach, you know, their Apple Watch or heart rate monitor to, a leaderboard. So you kind of get kind of similar audience theory where you get, you know, points based on how many minutes you're in the pink zone or orange zone or, yellow zones, etc..
00;01;51;25 - 00;02;12;23
Unknown
And so people competed there. It was interesting because at the moment, you know, peloton was taking off like a rocket ship. And so all of the traditional fitness manufacturers were kind of asking themselves the question, well, how do we compete? Right. And they were saying, well, we've they've got hardware, but they needed software and content. And so when they looked at us, they said, well, you can solve that problem for us.
00;02;12;26 - 00;02;33;17
Unknown
So we took investment or partnered with folks like Technogym and like fitness, Diko and Swami, most of some of whom I'm sure you never heard of. But like, they're basically all fitness companies who are basically asking themselves, you know, how do I compete with peloton? And that was both really beneficial at moments, but really kind of challenging at other moments.
00;02;33;17 - 00;03;01;01
Unknown
And so what you describe some of the more of the challenges? Well, I think corporations have different strategies or different, things that they're prioritizing. So if you are a manufacturing company and you prioritize manufacturing capabilities, that's the thing you're trying to optimize, not subscription revenue, whereas we are trying to optimize subscription revenue. And so kind of, you know, day one of a partnership, you're kind of you're aligned.
00;03;01;03 - 00;03;18;27
Unknown
But at the day of launch, you might not be aligned because you're like, oh, well, does subscription revenue mean that there will have fewer things to manufacture? And the answer is maybe. But the answer is, you know, our point of view was that you have more margin and contribution to a subscription. Long story short, is kind of, those things cause friction.
00;03;19;02 - 00;03;37;26
Unknown
Yeah, yeah, I sponsor my car, oil and gas. And I remember it was amazing to me. Company? Try not to name names. So we don't get anybody in trouble. Like you said, but the it manufacture pipe. And I remember trying to digitize all this stuff. Right. And so we were like, it would be awesome to get tracking traceability down to the individual joint.
00;03;37;28 - 00;04;04;23
Unknown
And how are we going to charge for it and what do you do and blah, blah. And ultimately, it was like a total dud in my mind, because it's a manufacturing company, you know, like we make pipe and, excuse me. And and so it is it's been fascinating, I guess, to think through like, like a, you know, an Amazon, which, I mean, while not a manufacturer, managed goods, and ultimately be like the Prime model, AWS model, everything that's, that's been built around that.
00;04;04;25 - 00;04;18;23
Unknown
I mean, I think it takes a special kind of person, like a Jeff Bezos to really think about that. But, it's amazing that they've been able to transition so well from selling books to. Yeah, yeah, the whole platform. Yeah. I mean, yes, I, I guess like this is my problem. I don't read a lot of news.
00;04;18;23 - 00;04;35;00
Unknown
And so every once in a while I get kind of situation where they're like, why didn't you read this article about Jeff Bezos that said X, Y, and Z? And I'm like, I should I should have yes, you're right, I should have. My understanding is he had this vision prior to the books, right? It was like, hey, this we start with books that are easy to ship and there's kind of flat rate and things.
00;04;35;02 - 00;04;57;03
Unknown
But, you know, I think the presumption was where can we expand to both to keep our, you know, commodity items so people know what they're receiving. Right. And over time, also, what about shipping? Because shipping of books is cheaper than shipping of, you know, non books because immediate media rates are a specific. Yeah lower rate. And but it's been interesting to watch them obviously kind of expand well beyond.
00;04;57;06 - 00;05;20;21
Unknown
Yeah. Their own inventory to other people's inventory to obviously things like AWS and everything else. Yeah, yeah. And then, you know, you've been through so many tech cycles. I'm like, yeah, he's anyways I'm all, yeah, basically. Yeah. You sure you want that much coffee now the, the, but like you started describing, you know, ed tech in 96 to to today.
00;05;20;21 - 00;05;45;27
Unknown
I mean, take me through these cycles because I feel like today everyone's a bit freaking out. They're like, man, this has never happened before. And I just don't have a point of reference. I mean, you certainly have some points of reference, I'm sure. But yeah, I think that what you're seeing, with the kind of, concern about new technologies, acceptance of new technologies and then adoption, you know, broadly of new technologies has happened, you know, multiple times in our lifetime.
00;05;45;27 - 00;06;02;16
Unknown
Right? So, the example I think you're referencing is when I came out of college, the internet was brand new, right? In fact, the first browser I used was the year I did my first internship. I was in 94. Right. So it's like, okay, that browsers there. You're kind of you say, oh, this just kind of cool. And you're kind of thinking about yourself.
00;06;02;23 - 00;06;19;02
Unknown
What is this for? I think it's Paul Graham who said, you know, a lot of new inventions look like toys at first. And that's kind of how the first browser looked. Right? But very quickly you start saying, oh, I can do ticketing on this, I can do, organization, I can do an internet, right? All these other kind of variations.
00;06;19;04 - 00;06;39;02
Unknown
And very quickly, obviously, the late through the late 90s, the potential huge impact of the internet became very clear with the.com bubble and eventually bust, of course, where people got very excited and, you know, maybe a little bit disenchanted as things kind of blew up and it kind of that cycle repeats, obviously with things like mobile or now obviously with AI.
00;06;39;04 - 00;07;05;19
Unknown
And so kind of the question is like, you know, how do people react and kind of the explosion of the potential, you know, the realization that kind of it might not be kind of quite as big as it is. And then slowly realizing it maybe could. And so, I think it's just a repeated kind of thing. And so right now I'm reminded of this, the cycle when I first showed my father in law, OpenAI or ChatGPT, he was like, oh, that's that's neat.
00;07;05;19 - 00;07;31;19
Unknown
But what was it? It was very first like, you know, the first month goes out or something. And he's like, that's pretty neat. But how will they make money? And I think to myself, it's like, you know, it's like, you know, it's a silly question. And I think, you know, he's he's not become a, you know, staunch devotee to ChatGPT uses it all the time, but same time, you know, as I said, you know, first the first versions of things look like toys that look silly and you can't quite get your head around what the maybe commercial impact will be until years later.
00;07;31;19 - 00;07;48;19
Unknown
And so, I think people are still getting their head around what the commercial impact will be for. I think it's people believe it will be big, but it might be bigger than that even. Yeah. I, I know I'm, I'm waiting especially for my portfolio companies. I'm waiting for the other shoe to drop where it's like, we're not going to subsidize all these tokens anymore.
00;07;48;19 - 00;08;15;02
Unknown
And and now, you know, the really start making money. I, I'm a bit terrified but okay. But that know comments on that too. I mean please please, please I mean like, you know, I think it's you can worry about folks like OpenAI and anthropic and kind of, you know, how much are they subsidizing? But you can also say, well, how does their cost compared to open models or local models and things like that, or Chinese models, etc.?
00;08;15;04 - 00;08;31;12
Unknown
And I think a way, the way I think about it, and I don't know if this is quite universal, is you kind of think about how much intelligence is my paying for and how much does that intelligence cost. And so the simple version I say is like, okay, imagine a eighth grader versus 12th grader versus a college student and PhD student, right?
00;08;31;12 - 00;08;49;20
Unknown
So today maybe a PhD student is expensive, but a 12th grader is affordable and an eighth grader is effectively free. Maybe it's some other variation, but in a year's time it will shift up one level, right? So next year, a 12th grader will be the equivalent of free and a PhD will be moderately expensive. And that continues to go forward.
00;08;49;20 - 00;09;14;00
Unknown
Now, you might always say I need the cutting edge person and I'm willing to pay for that premium. And that's, that's part of the cost question. The second part of the class question is actually the location ability store something. Right? So it's like, right now kind of that 12th grader might require to be in the cloud, but maybe eventually becomes on a powerful laptop, but maybe the year after that becomes an iPhone and eventually a watch or earbud.
00;09;14;02 - 00;09;32;26
Unknown
And so the question is, you know, these two things converging mean that in ten years time, let's say, which is pretty, you know, impossible amount of time in the future that everything will be effectively free and small. And so then you kind of say, well, that explodes the demand and utility of all of this stuff. So the concerns of your portfolio company are basically next to nothing.
00;09;32;26 - 00;09;50;03
Unknown
Yeah, yeah, I can it to like when we used to have to pay by the text message. Right. And and we're in the AOL CD time. Where's all good ten hours right. Yeah. So it's exactly yeah. I agree, I it is funny though, like the, the, you know, the human mindset of like any time the next model comes out.
00;09;50;03 - 00;10;14;24
Unknown
Opus 4.7 I have to use opus 4%. Why? You know, because I just think there's this this need, this desire to, like, always have the best. I mean, when you search something, you're searching for the best school site, the best toothbrush, you know, do you really need the best, you know, of everything, right? And yeah, I, I, I am curious how like, that dynamic will adjust because I'm already seeing in my portfolio companies are like, okay, let's let's test on the best model.
00;10;14;25 - 00;10;33;28
Unknown
Once we have it figured out, we'll move to an open source or a Chinese model, something a little bit cheaper. They can perform the same task because we've gone through the iterations of understanding what exactly it needs. And then but then moving forward, it's your to your point of around convergence. Like at what point do you stop seeking the best model because it's, it's good enough?
00;10;33;28 - 00;10;52;22
Unknown
Or is the competition always out there. And this boogeyman is always out there that requires you they need to do that. Well, I don't know. I think a bigger question is things that I don't yet know the answer to, which is, as you get beyond the PhD student to the next level and level after that, what does that unlock that we can't even in vision?
00;10;52;25 - 00;11;09;13
Unknown
I think that's probably more interesting. It's like as a way to think about what the next couple of years are. I mean, I'm solving kind of the basic questions of today's economics. But if you look on that, you say, oh, shoot, that means, you know, someone can do this task from A to Z, as opposed to just answer a small bit of the question.
00;11;09;16 - 00;11;31;13
Unknown
And that's another kind of, type of parameter people are talking about, which is how long can an AI model work independently without interruption? Or you know, in, just in, ingesting guidance from the human user. And if it goes from, you know, 2 minutes to 2 hours to two days, that becomes really incredibly powerful. If it can kind of work independently of the human interaction.
00;11;31;16 - 00;11;52;06
Unknown
That's really powerful. Right? So the foundation to the foundation model is just take over everything. I think it's to be determined. I think I'm not wise enough to know the answer. There. I mean, I like I'd like to think it's not right. I'd like to think there was a time where people thought Google would. And while they have a lot of that and taking over, I think I think they have a lot of power.
00;11;52;06 - 00;12;12;25
Unknown
But like, we use slack, we use AI, you know what I'm saying? Like there's there's a lot of there's this unbundling and bundling of tools and processes and workflows that happens. It just feels like right now everyone's trying to do both at the same time. Well, I think the I think that what you're going to see is probably emergence of expertise in different areas.
00;12;12;25 - 00;12;37;12
Unknown
Right. And so, you know, there's definitely a leapfrog, you know, kind of thing happening right now where one model is excellent at this in the next month, the next model's better at it. And you know, you know, continues ad nauseum. And I think the natural question is like, do they start? They all do, text and coding and, image and video.
00;12;37;15 - 00;12;54;27
Unknown
I think the answer is probably no. You can see it kind of where people are kind of cutting back. Obviously saw has been kind of reduced in terms of the emphasis for for OpenAI. You know, obviously kind of coding has become kind of a clear area of the frontier where they need to all win. So they're kind of investing heavily in those areas.
00;12;55;00 - 00;13;15;12
Unknown
I think a natural question is like, what is next? You know, things like legal and a variety of other, you know, medical, you know, frontiers are where people are starting to say there's really big potential here. I mean, this is not novel, but I'm just saying, like, those are kind of big areas which are still, largely unsolved or not or not solved.
00;13;15;12 - 00;13;32;05
Unknown
This color coding is, and so the question becomes, you know, who becomes the expert in each of those different verticals? I think the answer is definitely still out. I think right now there's just so much space. It's like if you're in the Wild West, you just go west. I don't really care if you're going northwest, west or southwest.
00;13;32;05 - 00;13;46;18
Unknown
You're just going west. And so it's huge, huge opportunity out there. Yeah. So I mean, moving on from studio now, I gave you, you know, a chance to start building on your own as you're building. Is this in the back of your mind or you're like, you know what? I'm just going to build because I need to learn.
00;13;46;18 - 00;14;10;05
Unknown
And and then from there we'll, you know, figure out how far west I get. Yeah. I mean, I think for me, when we sold studio and kind of my responsibilities there ended, I just knew I wanted to take time off for myself and time off for myself was a month of doing nothing and saying, I can't do this anymore.
00;14;10;07 - 00;14;34;04
Unknown
I need to do something. And I didn't need to know what the something was. But I said, I'm going to start, you know, exploring everything that's out there. And so, during my time off, I, you know, created a I recipe site and I travel site and I Q&A site, probably a half dozen different games. And all of that was either with a technical partner or solo kind of.
00;14;34;06 - 00;14;52;07
Unknown
And, you know, when people would ask me, why are you doing that? Is this your next company? And I'm like, no, I mean, my rest of this recipe site is clearly not a company. It's just me tinkering, trying to understand the capabilities of open AI, the capabilities of image generation, the capabilities of story generation, etc. following directions. See, when people don't give a shit about their data when it comes to food.
00;14;52;13 - 00;15;09;13
Unknown
And I imagine fitness too, right? It's like, yeah, sure, whatever. Take it. And but and it's and from my point of view at that, through all those different little projects was like about learning and trying and doing. And that's part of what I encourage anybody to do. I feel like it's very helpful at this moment to be unemployed.
00;15;09;13 - 00;15;29;13
Unknown
So you can kind of have, you know, lots of time to do stuff. And that's a, you know, humorous aside, I realized it's not really possible for most, but the, the intent of that journey was not just to learn, but also to explore kind of what I thought would be interesting. And I wanted to kind of, you know, have a big project to merge.
00;15;29;21 - 00;15;55;25
Unknown
And so the first big project that emerged was a project called smote. So Smote was, stood for Smart Notes. The general idea Smote was to record, campus lectures, turn them into, you know, obviously smart notes, but also kind of quizzes, flashcards and, you know, notes, a voice tutor that the student could interact, interact with, it turns out to very crowd space.
00;15;55;27 - 00;16;14;11
Unknown
Lots of different, recording, kind of, you know, alternatives out there, superior products in a market like notebook LM, you know, a variety of other ones. So it's like, you know, I was kind of going headlong into an area which I think was, you know, probably not as ripe for disruption as I'd hoped. And so kind of after a bit of exploration, I said, you know, is not my thing.
00;16;14;11 - 00;16;31;22
Unknown
And I decided at the end of last year to shut it down and no harm, no foul. I think it's very useful to kind of take, you know, swings kind of to explore stuff and then be okay if you miss. And then shortly after I was like, well, let me start scratching itches that are you know, more recent.
00;16;31;22 - 00;16;52;27
Unknown
Obviously, I haven't been a student for a long time. So that was a little bit less recent. Whereas, you know, my wife and I had just bought, a condo here in Houston. We had bought a condo in Miami. We had sold a house in upstate New York. We got through a bunch of real estate transactions, each one was, you know, annoying in its own particular way.
00;16;52;29 - 00;17;12;22
Unknown
Painful. And, you know, the truth is, there are many, many different, steps to a transaction. There's lack of clarity for you, the buyer, but also many steps that the, the agents all need to step through. And, you know, when I, we bought our condo here in Houston, my mother in law was our agent. And I love my mother in law, like anybody loves her mother in law.
00;17;12;24 - 00;17;32;03
Unknown
But she was not particularly active in the management of the deal. So my wife, who's really wants to be on top of things, was asking me to be on top of this. And I said, well, there's got to be a better way than, you know, trying to parse through these emails. And so my initial version was just to copy and paste them into ChatGPT and say, hey, help me make sense of this.
00;17;32;06 - 00;17;56;17
Unknown
And long story short, I started creating the new product called dealio. So dealio is kind of what I'm working on now. It's basically taking real estate and trying to, create playbooks for transactions that are easy to follow from start to finish. And some of you, I didn't know when I started this is that, you know, there are about 4 million transactions in the US, each year, but half of them use a transaction coordinator.
00;17;56;19 - 00;18;15;11
Unknown
That transaction coordinator is effectively a person who's responsible for gathering documents, establishing dates, providing communication amongst all the parties. I was like, well, that's kind of what, you know, I was trying to do with this one deal. Why do I try to do that with the help of AI? And that's kind of what Delia's been. And so that's been kind of like leading me to what I'm working on now.
00;18;15;11 - 00;18;42;18
Unknown
So we're about to launch. I'm pretty excited about that. And, one of the interesting things that's been happening in that period, and you and I talked about this is one point, which is, do I make dealio a tool versus provide the entire service from start to finish? And, you know, the big conversation in the world of AI and investment and kind of especially with, improving models, is to provide the entire service as opposed to just providing a tool.
00;18;42;21 - 00;19;08;07
Unknown
So what we are doing is going to be be the transaction coordinator for, for real estate agents. The nice thing is it's, you know, not just hopefully better because it's you can respond immediately. We can kind of gather and kind of have complete understanding of all the content you have share with us as the SEC, but it's also hopefully going to be cheaper because we're able to kind of automate, you know, maybe today we can automate 50% and over time we can automate 80 to 90%.
00;19;08;08 - 00;19;24;12
Unknown
Yeah. And so we're pretty excited about that. I love us. So I really I think it's starting in Covid, because I was bored, but also I realized, like, there's like a lot of dumb residential realtors and so, like, maybe I can do this myself. I got my real estate license, and and I just use it on our own.
00;19;24;12 - 00;19;48;15
Unknown
Our own properties and, you know, rental properties and whatnot. I, I have never used a transaction coordinator. So like, what type of realtor requires a transaction coordinator? Yeah, I mean, I think if you think about the skill set of most transaction coordinator or sorry, most, real estate agents, excuse me, is they're excellent salespeople. They are people who are able to get a deal done and help negotiate negotiate the deal.
00;19;48;18 - 00;20;12;03
Unknown
And when they're done with that, it's it takes a special type of person to switch and say, now, let me get into the minutia of the deal and go through every document, go through every day. And so for the folks who say, I want to stay really focused on selling, and I don't want to worry about the minutia of the deal, that's the person who's saying, you know, I'm going to do that and have the transaction coordinator do those those deals.
00;20;12;05 - 00;20;32;02
Unknown
Got it. And then, but I think it's also like, you know, a lot of people need to ask themselves, where is their expertise? And so, you know, maybe you might be an expert in kind of organizing documents and coordinating dates and all that stuff. No, but and, and you start realizing, well, for most people, that's actually kind of a little bit of a headache.
00;20;32;02 - 00;20;57;22
Unknown
It's a little bit of a nuisance. And, you know, quite frankly, a robot or an AI model can maybe do it better than you as a human. And, you know, you can do it with fewer complaints and kind of faster response and greater accuracy and hopefully lower cost. And so, because of that, you know, if you if you kind of think about a transaction where you might be making, you know, several thousand dollars, you're saying, can I afford 3 or 4 or $500 for transaction coordinator?
00;20;57;27 - 00;21;15;14
Unknown
You might say, oh, it's a it's an easy cost to, to take on to kind of move on to the next. Yeah, yeah, yeah. Okay. So that's what I was is like a higher volume obviously justify something like this in particular a realtor that does a great job selling things like this personal injury attorneys that you see billboards everywhere for like a lot of them are just great marketers.
00;21;15;14 - 00;21;31;25
Unknown
And then they just third party out the rest of the work. That's right. Okay. Great. In a log there. And, and, or for example, and same staying within the same industry, property management. Right. I would my friends, especially when I had rented a lot more rental properties. I was like, is a property manager worth or not?
00;21;31;27 - 00;21;50;21
Unknown
And everything I think in life is just on this range of comfortability to profitability. Right. And so the what, what I'm hearing though, is that like now with, with dealio and with other tools like we, we might be shifting this entire range right where you can stay comfortable and profitable in ways that you couldn't before. That's probably true.
00;21;50;21 - 00;22;15;12
Unknown
I think it's, it's not just those things, but also probably capabilities, which is like, you know, like I think to myself, what are the things that kind of are unlocked that, you know, we're not possible before. So when I think about dealio, I think not just can I this is kind of goes to my product philosophy, which is you kind of build for single player, but you have a vision for multiplayer.
00;22;15;15 - 00;22;36;07
Unknown
And multiplayer is often where people are never able to participate because just doing single player, you know, in a standalone is already hard enough. So what I mean by that is, you know, if you're doing 20 deals a year, like some agents do, like you're, you're just trying to keep up with doing, you know, those 20 deals, if you can all of a sudden have it.
00;22;36;07 - 00;22;56;03
Unknown
So you have a dashboard of for all of that, you can have it so that, you know, the broker understands it, the team understands, you know, where things are from. You know, this is 20% done to this is 80% done to. These are the the blockers. That becomes really interesting from the you know, you know, your manager kind of having an insight into that or you're having insight into your whole team or where they all are.
00;22;56;06 - 00;23;12;15
Unknown
And all that stuff is part of the unlock, not just the comfort versus the profitable it ends up being kind of, greater insight and hopefully just greater growth overall. Yeah, but the reward for good work is more work. Right? I have to mention this. Like, that's, like, it can be overwhelming. The amount of stuff you can do.
00;23;12;15 - 00;23;34;14
Unknown
Right. And, and and I like real estate tech, this space. And I feel like there's so much happening, that, first of all, I'm glad your brain is trying to solve these problems because it just feels like there hasn't been enough movement. You know, it's like people are still stagnant in their legacy, their legacy processes, their legacy tools, their legacy structure.
00;23;34;16 - 00;24;04;10
Unknown
I, I'm I'm fascinating by how I mean, in your discovery process, you may have seen, okay, that the adoption of something like this and did that, did that, like I guess infer your decision to maybe just be the service provider? Well, I think it's, you know, the natural question becomes what is what can be adopted, right? So if you go to a transaction coordinator and say, hey, you're making 300 up to the good ones, make up to thousand dollars per deal.
00;24;04;13 - 00;24;19;20
Unknown
And I want to charge you to use my tool. They're like, no, no, I you pay for doing this thing. I don't want to have any of my margin cut away. But if you say, well, let me take up another level and say it to the actual person who's the agent and say, okay, you're making $10,000 for this deal.
00;24;19;23 - 00;24;43;19
Unknown
You know, they're, you know, yeah. Their sensitivity to $300, $40, $500 is much lower. Right. And so, and so it ends up being kind of natural to kind of go that route. And, you know, one of the things we haven't really touched on, but it's been, you know, part of, a lot of discussion about AI investment is when you take on the services revenue versus just the tool revenue, your your Tam can be bigger.
00;24;43;19 - 00;25;02;12
Unknown
Your, you know, as AI models improve, you can kind of like do things more. Automating your margins can improve. And so all those things are part of the, the approach. But I think underlying all of it, which I think you touched on, is the adoption, which is like how do you simplify adoption? And if someone says, I do a deal, I hire transaction coordinator.
00;25;02;14 - 00;25;19;14
Unknown
That's a lot easier than I do a deal. I hire a transaction coordinator and hope that they use this tool. Right. And so like, we're cutting that layer out and, you know, going directly to serve that. Yeah. Do you foresee though eventually, I mean, as you outpace these other transaction coordinators than being like, what's going on over there?
00;25;19;14 - 00;25;33;10
Unknown
What's, what why is the thing doing in crushing this so much? Well, I think it's, it's possible. I think kind of, you know, the market is so massive. I mean, there are 4 million deals a year that are doing it, you know, in the US alone. And so the question is like, how many do I need to become big?
00;25;33;10 - 00;25;52;11
Unknown
It's like it's yeah, it's a small, small fraction of that. Right. And so I think that it's going to be a little bit of time before we're even. Yeah. People realize we're here, right? If I'm being honest. Yeah. And if we're at $10,000 a year, that's actually quite sizable, right? It's actually pretty big business. So you're saying okay, well let me go to that first before I try too.
00;25;52;13 - 00;26;15;23
Unknown
Well, this drives me to another philosophical question that I've run into with a lot of technical founders right now. Is that, like, again, I started in the technical space, in the tech space, in technical tech space or 5 or 6 years ago. You know, learn how to spell VC maybe three years ago. And so what, you know, like what what I, what I see is, is very relative.
00;26;15;25 - 00;26;32;10
Unknown
I guess what I, what I remember hearing when I first started was, okay, things have to be as scalable as quickly, as big as possible. And that's I mean, that is the VC game all to me, right? The power law requires you to do that. What I'm hearing more from from technical founders now is I you know, you know what?
00;26;32;10 - 00;26;47;20
Unknown
I'm fine with the lifestyle business. I can't wait to just build a profitable software. I don't think that it's what I'm describing as a lifestyle business at all. I think it's it's frankly, it can be massive. And I think it's useful to kind of just do a little bit of back of the envelope math to figure out how massive it can be.
00;26;47;20 - 00;27;12;06
Unknown
Right? So if you if, if I presume that, an average deal for us starts at 300, maybe average at $500 per deal, then you have ten deals is 5100. Deals is 50,000. Was it, a thousand is was that so. Right. And so like you start kind of getting to very big numbers very quickly. Right. And so you start saying, well, shoot, if I can kind of do $5 million a year, that's $60 million.
00;27;12;09 - 00;27;30;13
Unknown
That's right. $5 million a month, that's $60 million a year. That's a very massive business. And and, and that's kind of what I would consider layer one. Right. So then you say, well, what's layer two? What's like what about the broker view? Okay. The broker looks at all their 15 people on their team and they want to see all that.
00;27;30;15 - 00;27;50;06
Unknown
What what does that look like. Or what about legal advice. And we're not trying to do that obviously at all. It's a, it's it's a dangerous area. But like, can you eventually kind of provide legal guidance or can you provide kind of other types of coordination or other types of layers? And I think it's very useful to, to be as narrow as you can to start.
00;27;50;06 - 00;28;10;29
Unknown
So you don't confuse the message. But I think that if you kind of in this exercise, looking on the horizon, there's a lot of things that can layer on top of this basic service. Yeah. So yeah, that's cool. How did you end up in Houston? The love of my life, my wife Tiffany, she grew up here, and so kind of, I guess it was your wife and you were in my.
00;28;10;29 - 00;28;39;02
Unknown
I love, you know, but as, you know, we were, you know, at a moment where we were taking our respective sabbaticals, we kind of said, well, you know, every for the last five years, every November and December we've spent with her family here in Houston, and we kind of eventually said, you know, instead of, staying on a in-laws, you know, extra bedroom in their extra bedroom, well, maybe actually get a condo and stay here for longer.
00;28;39;04 - 00;29;00;21
Unknown
So right now we're splitting our time between New York and Houston. I don't know if that will be forever, but it's good for right now. I think that kind of part of me misses kind of the the tech scenes in New York or even California and elsewhere. But at the same time, the biggest differences. Well, I kind of joke about this a little bit, but I went to an AI event.
00;29;00;24 - 00;29;20;13
Unknown
I think I may have told you about this, but and, you know, I was expecting them to talk about Open Claw or, you know, a variety of other kind of really innovative new things. And instead, the words I heard were about change management, which is, you know, effectively, how do you take an organization of a thousand people and have them use ChatGPT for the first time?
00;29;20;15 - 00;29;43;02
Unknown
And I'm not try to discount that. I just that doesn't seem like innovation to me. And so, when you go into the, the places where innovation, I think is happening more actively, you are talking about like what is next? And so, you know, Gary Tannen talking about his G. Stack and you have people talking about the, you know, harness versus the, the models you have.
00;29;43;02 - 00;30;07;18
Unknown
All these are kind of interesting, I think, conversations about what is next. And what's around the corner. What are the second order effects? I think in order to, to, either, you know, benefit from that, I think you need to either be in that, in those communities directly and talking to the people directly, or you need to really embed yourself in the communities where I think right now that's probably Twitter.
00;30;07;21 - 00;30;27;23
Unknown
And I think that, you know, like the reality is you need to kind of know what is happening in order to even have a possibility to know what's next. Yeah, yeah, yeah. No, it does feel like the flywheel just spins faster there. Definitely California, New York, other places like that, even Atlanta, where you have this, like, sneaky, incredible tech ecosystem.
00;30;27;28 - 00;30;43;04
Unknown
Maybe, maybe by function of that, like you've had successful tech companies that have been recycled and and, you know, seeing is believing to a certain degree. And they're like, okay, well, if someone can do that, then maybe I can do it too. Versus here. That's incredibly powerful. I mean, that's frankly the very first startup I was ever part of.
00;30;43;06 - 00;31;01;02
Unknown
The founder was a colleague of mine from McKinsey. Very smart guy, but more importantly, just willing to take a risk. And, you know, when I worked with him, I saw that he was willing to take a risk. And it made me say, well, if he can do it, maybe I can. And it wasn't that he was a superhuman.
00;31;01;04 - 00;31;21;05
Unknown
He's a great guy. I don't it's very true. But like, it made you say, well, shoot, like you need role models that are up close and personal and you can relate to to say, maybe I can jump off the bridge too and have a parachute to go get that right. And, and, and, and in places like San Francisco and in pockets in New York, you have those entrepreneurs around you who are doing it.
00;31;21;05 - 00;31;36;19
Unknown
And so you saying, well, you know, if they're not as crazy as I think, and so maybe I can do it too. You sound like my therapist now. I, I just, I mean, hopefully that you other. No, no, no, I'm in a, in a great way. Meaning like, I, I, you know, I never really had room.
00;31;36;21 - 00;31;55;07
Unknown
My, my, my family are doctors, which is God bless them. But I just like it wasn't for me. And and then oil and gas. I just never really saw anybody that I was like, man, I want to be that person. Didn't matter how wealthy they were. You know, went through my trials and tribulations through startup life and, and I, you know, I'm, I, I'm still looking around.
00;31;55;07 - 00;32;16;17
Unknown
This wasn't meant to be about my, my psychological issues, but but it is it is, I think, a good embodiment of, like, where the ecosystem sits in Houston and other places like Houston, where I think it's easier to look at someone that's had success in a brick and mortar SMB or in oil and gas or as a in the medical profession and say, okay, I can go do that.
00;32;16;17 - 00;32;35;26
Unknown
And almost blinded to the opportunity that's out there, tying it in, landing the plane a little bit around a technical conversation now with I, I'm, I am fascinated by the ability of like, people to remove those blinders and be like, oh my gosh, I've become a subject matter expert over ten, 20, 30 years of this thing. What can I do now?
00;32;35;28 - 00;32;52;02
Unknown
You know, and, and, and I love that you get the chance to explore that a little bit. I am curious why, I guess dealio. And then the real estate market versus some of the other things that you did that maybe you had more expertise in. Right? Yeah. I think the problem is I'm not really an expert in anything.
00;32;52;04 - 00;33;13;05
Unknown
And I'm I'm kind of somewhat joking. I mean, I've done, you know, half dozen real estate deals. I probably done 50 investment deals. And, you know, the truth behind dealio is that we initially started saying, let's do all deals. And the problem with that goes to my consciousness for my studio was, when you try to boil the ocean, it's like, it's not going to work, right?
00;33;13;05 - 00;33;36;07
Unknown
So you say, well, how do I get more narrow? So a studio, what we did was say we actually my favorite moment from the original studio is we had a list of all of the, types of workouts you could take from running to, on treadmill to bikes to ellipticals to everything. And my co-founder, you know, did something really brilliant, which is you walked up to the board and crossed out nine of the ten.
00;33;36;09 - 00;33;52;23
Unknown
And so what if we just do this one? And it was like was shocking because I felt like, why would we limit ourselves? And the reason to limit yourself is to be so focused. Right. And we always knew that it's a lot easier to go from 1 to 10 than from 10 to 1. And so it's not sort of similar with Delia.
00;33;52;23 - 00;34;19;16
Unknown
Right? Which is to say there are millions of types of deals, right? Everything from, you know, obviously a real estate deal. But, you know, we've been involved in VC deals or angel deals or M&A deals or, you know, elbows or whatever else. And they're all, you know, they rhyme, but they're slightly different flavors of the same thing. And so I said, in order to have a clearer go to market and to have a clearer kind of product set, I wanted to pick one sliver.
00;34;19;18 - 00;34;38;03
Unknown
And that's not to say that the other slivers will not eventually be there, but if I if I don't start with that one sliver, it becomes too confusing to know what we're doing. To know we're talking about, you know, HOA and title and loans and all these other pieces of the puzzle. But because we're so focused on that, there's already enough, you know, yeah, complexity there.
00;34;38;06 - 00;34;59;27
Unknown
Let's just focus on that. That's that's become more special. Nathaniel, I like the ability to do so much more now, I'm sure has caused people. And just like the short term satisfaction ism that we've grown up with now, as a millennial, I feel like it's so unique to have that level of focus. And, well, one, it's exciting, but but two, I think it's becoming it's funny.
00;34;59;27 - 00;35;19;20
Unknown
I think it's becoming a superpower. I mean, like as you've engaged with more and more people, have you felt like you you notice that more or it's just like so innate to who you are being focused? Yeah, I think that it's a different well, I mean, if you show anybody anything that I think you kind of want in, a certain type of reaction, which is.
00;35;19;22 - 00;35;38;22
Unknown
Yes, and could to do this and I think that's both, an, an indicator that their imagination is going to the next thing, which is great, but it also takes, it can naturally pull you away from your focus. Yeah. And I think good entrepreneurs can kind of take that feedback, compartmentalize it, organize it so they can use it in the future.
00;35;38;22 - 00;36;07;23
Unknown
Yeah, but not have it, you know, redirect them prematurely. And, I think it's, you know, but I think it's, it's, it's you kind of want people to start thinking about the potential of what this could become, because that's like, you know, going over what I call the imagination problem. The imagination problem is, you know, when you show someone something that doesn't quite work and is a little bit broken and is like, you know, has the future is not quite fully pull through.
00;36;07;25 - 00;36;24;13
Unknown
You know, most people when they look at they go, oh, it's broken, right? You're like, you stupid idiot. Like, no, of course it's broken, cause we're not. We haven't, you know, put that button there. It's not. Call it the right way. It's not lined up. And you want people to be able to kind of say, oh, I imagine if a button works, it would do this magical thing.
00;36;24;15 - 00;36;44;25
Unknown
And, but to kind of get it to it does the magical thing. It takes, you know, months time, sometimes a work, right, maybe a year or so in some cases. And, but you want to find around you as people kind of get over that imagination hump rejection problem and start realize, but, you know, potential of the thing that you're you're presenting even before it's done.
00;36;44;28 - 00;37;03;10
Unknown
And that kind of goes to the problem with focus, though, because when you once you get there, they're like, oh, great. Now you've got to this point, what about this next thing you're like, let's just focus on this and then we'll get to that next. How do you transition though? How do you get at what point you say, okay, I've, I've reached a certain threshold that I've wanted to reach while focusing on the, on this element.
00;37;03;10 - 00;37;26;00
Unknown
And there therefore now I need to do that. I don't know if I have a good answer for that one, but the, I think the, you know, the in conversations you're hearing, you hear things over and over again. Right. And so the question, of course, is how do I not just compartmentalize them and organize them, but eventually say, okay, this has been said ten times.
00;37;26;06 - 00;37;47;00
Unknown
This has been said once. Let me ignore this. Let me kind of focus here. You know, in our life with the studio, we started realizing that working with hardware manufacturers was going to be important for, you know, some of our strategic future. And we had people on the other side saying, oh, an Apple Watch is dumb because it might be a metal band and people never wear that.
00;37;47;03 - 00;38;05;20
Unknown
And I was like, you know, so you kind of have these two different sides to this where you kind of say, which piece of feedback do I listen to and which part of feedback do I ignore? And, I think it's, you know, the good product focus leaders are going to focus on, you know, that, you know, community feedback and kind of, you know, sift through it.
00;38;05;22 - 00;38;24;06
Unknown
Yeah, I've struggled personally with this. This is a bit of a selfish question, which is like, I, I, I find that that, you know, when you think about a disruptive company or a disruptive tool, whatever the product, you think, okay, if I just ask people what they want, they might just say a faster horse. But I want to build a car.
00;38;24;09 - 00;38;43;07
Unknown
Yet most people, organically or just because they found success with this process in the past, we'll go ask and say, okay, well, I've seen this ten times. And, you know, I'm curious how you balance that, asking customers for what they want versus just kind of setting a vision yourself. I think it's,
00;38;43;09 - 00;39;05;05
Unknown
I mean, I think the most magical products are when people are presented with things that they didn't realize they wanted. And so, and, you know, I think like opening hours is a great example, but he's a great example where you type something in and like, you're like, oh, and I didn't realize that could ever happen when the first time you ever used it, you probably had that experience yourself.
00;39;05;08 - 00;39;28;21
Unknown
And it goes to what does it do for your, reaction, your reactions? And oh, but can you do this and can you do that? Can you do this other thing? And I think the, you want products that kind of like, you know, reveal themselves to you and can have how impactful they can be. I think you were never asking for a chat, a chat bot that was kind of going to answer questions that you were curious about.
00;39;28;24 - 00;39;43;27
Unknown
And I don't know, I mean, I think, but I don't know if I have a kind of a good solution for how do you kind of discover products, because I think the problem is most people don't have that capability. Right? Right. Yeah. And I mean, maybe it's on balance isn't the right word. Right? But it's like, how do you how do you identify that?
00;39;43;29 - 00;40;05;25
Unknown
And maybe to your point, maybe over time, as you have engage more, you get this your own eureka moments or you're like, yeah, absolutely. I should be building in this direction because that's where the, the, you know, the puck is going. Well, I mean, going back to like, you think about one of the things that we talk about a lot is in our organization is don't care about token cost.
00;40;05;28 - 00;40;33;24
Unknown
Obviously, we care about unit economics, we care about profitability, all those things. The reason I'm ignoring, the cost today is because I think that the cost, parameters will be different in six months, in a year and two years especially. Right. And so kind of by the time we are an important company, it will be intimately different, you know, number and if we have been only working on the ones that were profitable for us on day one, you won't be capable in year two.
00;40;33;26 - 00;40;50;07
Unknown
This is this is the the like, short term, long term tension of a lot of things, but in particular venture capital. It feels like, okay, I need to show that my investments are doing great well in this curve so that I can go raise another fund, when in reality, maybe what what they're doing isn't effective in the long term.
00;40;50;07 - 00;41;23;07
Unknown
And some of the, you know, some of the best companies, the best returners are the ones that like, hey, you don't see anything or they didn't grow in the way you wanted them to grow, or the first 2 or 3 or 4 or 5 years and then boom, exponentially. Because they've built this like bamboo root system that now rocket ships, I, I, I've, I've been terrified coming from oil and gas coming from real estate where even the hospitality world where if you don't make money, you don't get to work the next month, you know, into a world where it's like, no, we're going to lose money for a long, long period of
00;41;23;07 - 00;41;43;05
Unknown
time. Before you do this, how you've always been in tech, presumably like us. You sitting on the other side of that equation, how do you view the flip side where it's like you identify all these profitable businesses, especially here in Houston. You've got a lot of profitable businesses that are just not tech enabled, not tech forward.
00;41;43;07 - 00;42;08;16
Unknown
Does it? Does that affect how you think about your build for for dealio or otherwise? Well, I guess the when I look around and see profitable, profitable businesses, I actually have a little bit of jealousy. I, you know, my the co-founder of my first company, we used to joke that, you know, he had he had started a nonprofit, before we worked together.
00;42;08;16 - 00;42;29;03
Unknown
And I used to joke people that was intentional. Whereas the company we worked on together was nonprofit accidentally. And, now with OpenAI, you can do nonprofit and IPO. Yeah. You. But I think it's, you know, and, you know, obviously Amazon was famous for, I know, 20 years or something not being profitable. And that was by design.
00;42;29;03 - 00;42;46;15
Unknown
And I think the important part is, you know, what are you invested in and what are you, you know, able to use the money in the organization if you have a profitable bakery is probably because you don't have like some sort of novel thing to go put the money into. Right. So therefore, okay, I'm gonna take the money out.
00;42;46;17 - 00;43;06;28
Unknown
But a technology company that's taking money out means they have no new ideas. Right? And so that's probably a bad sign as opposed to a good sign of profitability. And I think, you know, it's been really interesting, I think for many years folks like Apple and Google and a lot of the other big, you know, seven big companies or whatever have been stockpiling this cash.
00;43;07;01 - 00;43;24;29
Unknown
And people be like, why are they stockpiling it? And it's actually very interesting. It's all of a sudden now in this moment when you kind of literally do need piles and piles of cash, many of them are able to say, well, now I've got piles of cash. I can go apply it to an area that I think will power my next generation of technology.
00;43;25;01 - 00;43;42;23
Unknown
And so it's kind of proving, you know, that, like, their profitability is now powering the next generation. But I think kind of when it comes to, you know, me as an entrepreneur, I think of it's like, you know, I clearly want to think about, you know, how do we get to profitable unit economics at some point?
00;43;42;26 - 00;44;01;08
Unknown
But I think, you know, what we're doing right now is investing in the, the base in order to kind of have something repeatable and scalable because, you know, if I can do ten deals, no one gives a crap. If I can do 10,000, that's a different story. And so I'm trying to figure out how do I do 10,000, and get the infrastructure in place to do that?
00;44;01;11 - 00;44;17;12
Unknown
Easily as opposed to like manually. Yeah. Well, soon you'll be making a ton of money on your bets, right? Yeah. I mean, yeah. So tell me more. I know we talked about it, all right. Yeah, I know I want to hear more about the yeah the poly Kelce arbitrage I mean I think that's really fascinating.
00;44;17;15 - 00;44;43;10
Unknown
One of the the things that's possible right now with whether you know your your your method, your replit or codex or cloud code or anything else, to develop because it lets you develop and try things that you might not otherwise. My technical background, I programed in 5 or 6 programing languages when I was in college, then went off to McKinsey, didn't program again for 20 years, and now I'm programing again.
00;44;43;10 - 00;45;03;08
Unknown
I mean, granted, I'm not going in, you know, readjusting where the semicolon goes. I'm focused on when you're talking to an AI agent and figuring out how to use APIs and a variety of other things. But I'm not, you know, coding in the traditional sense. But what's allowing me to do is to say I have a whole set of ideas and things I'm curious about, but I want to explore.
00;45;03;08 - 00;45;19;22
Unknown
Let me go explore them. So I think you were referencing was, you know, everybody in their mother's talking about, oh, we'll look at Polly and look at Kalki and let's go kind of explore that. So, and I was joking earlier, is your mother actually talking about this? My mother's passed away, so. Yeah. No. Yeah, that's,
00;45;19;24 - 00;45;44;09
Unknown
But no, she's not. Certainly not talking. Yeah, but the, but what's fascinating is like you start saying, well, how do I kind of build on Kalki or Poly Market to kind of, you know, explore different types of embedding strategies and approaches to, to attempt to win money. And I was joking earlier, all of my coding buddies so far have allowed me to, you know, different slow ways to lose hundreds of dollars.
00;45;44;11 - 00;46;03;17
Unknown
But I think that the next one is going to be profitable. So it's getting slower and slower, right? It's getting slower. Slower. I'm kind of losing money slower and slower. Are you willing to talk a little bit more granularly on how you're doing it? Yeah, of course. I mean, like, programing is really relatively straightforward because there are APIs for both, Gaussian and Poly Market.
00;46;03;20 - 00;46;31;03
Unknown
I'm trying to give like the different, set of, of wagers that I did. So I, I'll tell you, one of my favorite was I was focused on underdogs. I looked at 100 years of baseball data and I said, in baseball games, if a team is down by one run or two runs or three runs in the fourth, fifth, sixth, seventh, eighth, ninth inning, what is their percentage chance of winning?
00;46;31;05 - 00;46;51;05
Unknown
And I don't remember the numbers exactly. I'll tell them ahead, but let's just say if you're down by two runs in the seventh inning, historically, it's about 15% chance that you still would win. Which seems, you know, that's relatively high. So then I what I was doing was looking at what the odds were on Poly Market, and when there were roughly half that.
00;46;51;05 - 00;47;11;05
Unknown
So let's say it's 50%. If there were 8% or lower, I would make an automated wager on that. You know, I was very small, but, and, you know, as I was saying earlier, they often lost, despite the fact that the historical odds tell me I should win. And I think, you know, if you think about that implementation, I just described, it's incredibly basic, right?
00;47;11;12 - 00;47;32;20
Unknown
All I'm saying is, if the odds are there, bet. Yeah. Without taking into consideration all the other factors like is it a good team or bad team, is there, you know, is there star relief pitcher in there or not. Right. Is the heart of the lineup coming up or not. So is a thousand other things to layer in there that you can kind of hopefully improve that those odds to identify one is actually a good bet, a bad bet.
00;47;32;23 - 00;47;52;15
Unknown
So what I was doing was, I would say really kind of basic. Yeah. But over a long extended period time. Well, the problem is that the only time that the odds get that low are when the, or possibly when the, the market knows better than me. That's that's the fear. Right. And so they do say that Vegas was not built on the backs of winners.
00;47;52;15 - 00;48;13;18
Unknown
Right. Exactly as they used to say. And, invested. But where the customers got and so it's like the, same concept. Right. So, you know, I'm trying to think about kind of alternative approaches and alternative, betting methods that kind of work. And I think, I have some new ideas I'm not yet to yet reveal, but like, I think I have promise.
00;48;13;18 - 00;48;37;04
Unknown
And so is baseball the right the, like the easiest sport to kind of do this with? I mean, I don't know if it is. I think I mean, beside insider knowledge. No, I think that in some way you're looking for where there is, frequent activity. And so, you know, baseball might be 15 games a day. So it's actually it's somewhat frequent.
00;48;37;06 - 00;48;55;00
Unknown
But, you know, you can think about what else is more frequent, like, so how she has 15 minute crypto markets, probably market has five minute crypto markets, some with relatively good volume. Maybe there's something interesting there. I did do an experiment there trying to like look at the price of Bitcoin again. It was never quite right.
00;48;55;00 - 00;49;09;12
Unknown
And so you know also not profitable. Yeah. But like it's and and it's really easy to kind of say well that's dumb right. And of course it is kind of dumb to lose money. But the goal is in my mind is to experiment, to figure out, okay, well, how do I fix it? So the next time it's not losing money?
00;49;09;12 - 00;49;29;17
Unknown
Yeah, yeah, yeah. And you can do that. Then all of a sudden it becomes really quite a nice thing that just let sitting make you money. So what, how do you stay ahead of a poly marketer? Koski there's I'm saying is like those those odds are built based on how the betting activity goes on to say what they think will happen.
00;49;29;17 - 00;49;48;08
Unknown
Is that fair? Yeah. I mean, it's basically people are buying and selling, a yes or no outcome for specific activity. Right? So let's just say it's the Yankees versus the Mets, right? You know, will the Yankees win or the Mets win? And, actually, the very first version of what I did was arbitrage.
00;49;48;11 - 00;50;12;08
Unknown
So, first let me set the stage. So you have, you either position the Yankees or the Mets, you pay a certain number of sets. So let's just say it's, the Mets for $0.25. And if they win, you get a dollar or the Yankees for $0.75. And if they win, you get a dollar. Either way, you get a dollar and you're deciding how much you're willing to pay for that dollar.
00;50;12;10 - 00;50;29;07
Unknown
And the arbitrage I was referencing earlier is imagine you can buy the Yankees for $0.75, or the Mets for $0.20. That equals up to $0.95. If you bought both sides, well, then in that case, you buy both sides, and then you just wait for the game to complete and you collect your dollar. Hey, you're only making $0.05, but maybe you can do it.
00;50;29;10 - 00;50;51;14
Unknown
You know, a hundred bets of that and you get your five bucks. And so with this taught me was, oh, that's really interesting. But two things matter. Speed matters. And so kind of if you're using a VPN in Tokyo, it's which I was at one point that, you know, doesn't work very well because it's, you know, taking too many latency is too high.
00;50;51;16 - 00;51;09;11
Unknown
Or imagine a one side Phillies, but the other side doesn't fill. Then you have another problem, which is you have now one sided bet, which is you have no insight into whether the Yankees or Mets are going to win. And all of a sudden you have a bet on that game. It's like it's kind of silly. Or imagine it's a partial fill and all these other kind of little variations.
00;51;09;13 - 00;51;29;09
Unknown
And if you think about all of those were effectively lessons in, okay, how do I deal with latency, how do I deal with partial fills, how do I deal with no how do I deal with, slippage? Right. So the bet is $0.25. By the time I fill it, it's $0.30. And so like all of those are effectively micro lessons in how to trade in this environment.
00;51;29;12 - 00;51;47;07
Unknown
And so, you know, I look at it like, okay, I spent a couple hundred bucks learning these lessons. Right. And so it's, you know, my wife says that's coper and that's, you know. Yeah, yeah. No, no, no, but when, when our, when our, you know, a certain percentage of our GDP is built on, betting, I know who to call.
00;51;47;07 - 00;52;08;23
Unknown
And I will say, look, I'm not really a, I guess I'm, I enjoy playing poker, but I'm not. I'm not a better in the traditional sense where I'm saying, oh, I really have a good feeling about this team. I, I all the wagers on describing have to do with math and trying to figure out kind of what is a market edge versus, insight into the specific team or whatever it might be.
00;52;08;23 - 00;52;25;29
Unknown
You're a you're a calm, collected dude, actually, that maybe think what what has been like your most heated AI argument in the last few months? I think it has a lot to do with,
00;52;26;01 - 00;52;52;25
Unknown
Well, I mentioned I love my wife, right. And so I like where this is go. No, but, I am, very optimistic about the future of I am I think my wife is conservative or or just cautious. And I think it's it's it probably represents a vast, you know, a large portion of the world. Right. Which is, you know, there are definitely a lot of people who are very excited about where this could go.
00;52;52;27 - 00;53;19;16
Unknown
There's a lot of fear and uncertainty right and down. Right. And, you know, the impact is unclear. Will people lose jobs? Well, kind of, you know, things like mythos kind of, you know, create huge security problems. Will there be a variety of other kind of concerns that kind of we don't even realize could happen? And, you know, I was recently, recently watching a podcast with Fred Wilson, for Mean Square Ventures, talk about this.
00;53;19;16 - 00;53;40;05
Unknown
And he was basically saying, you know, look, it's not that I, I'm not ignorant to these things. It's more that where do I choose to focus my effort. And so in the arguments I have with my wife about this, it's like, it's not that I'm ignorant to these possible negative downsides, it's more that I need this because of my psycho psychological approach.
00;53;40;11 - 00;54;00;16
Unknown
I need to focus on the potential and the opportunity versus like protecting the downside. And I do believe that, there's huge opportunity in protecting us from those downsides. It just not for me. I want to focus on. Yeah, yeah, yeah. I think generally, I mean, maybe this is an entrepreneurial trade. And so I've just surrounded myself with this and, and I live in a bubble, so to speak.
00;54;00;18 - 00;54;19;15
Unknown
But I do feel like optimism far outweighs the pessimism, of what's possible or what will be possible with AI. I, I like I was, I was talking to actually a, a CEO of, of a manufacturing company, the speaker company. And he was like, I'm now teaching my kids a lot more around the humanities.
00;54;19;15 - 00;54;42;23
Unknown
He's like, because I believe that I will have an optimistic result, but only with, you know, humanities at heart, you know, and keeping those principles in mind. And I thought I found that fascinating. I was like, you know, said differently. So and one of our other partners, it was like maybe in the next five years, the most, the largest amount of, like, graduates are taking software engineer classes or AI engineer or whatever, whatever it is.
00;54;42;23 - 00;55;04;22
Unknown
But like in ten years, I think it'll be more philosophy. Right. And and spirituality and things like that. I as a thought experiment, I find it I find it interesting because I have to imagine that with every technical wave, people, when they're uncomfortable, will revert back to what makes them comfortable. And and whether you might call it a crutch or whatever.
00;55;04;24 - 00;55;26;21
Unknown
I, I think the, you know, the human element of all this is what makes me feel most optimistic about where this is going. And, and I think what gives people a lot more comfort in what's possible. Well, react to that a little bit, which is, you know, I grew up in a family where my mother was a reverend, and my father was a Ph.D. in biochemistry.
00;55;26;23 - 00;55;49;29
Unknown
My brother has a Ph.D. as well. So I was a dumb one in the family. But the, you know, it ended up this constant conversations about faith versus science or faith versus proof. Right? And I always thought that one of the really interesting distinctions was, how do we answer questions we don't have answers to? And oftentimes that's where faith steps in and tries to help us right?
00;55;49;29 - 00;56;11;02
Unknown
Which is where do I come from? Where do I go when I die? How should I behave while I'm here? You can argue like different faiths. I'm not trying to do that. I'm mostly kind of thinking about this concept of, you know, when do we have proof of something versus, you know, hope for something or lack of, substantial proof or not enough prefer not acceptable proof to somebody.
00;56;11;04 - 00;56;33;04
Unknown
And, I think that becomes really interesting and kind of goes to your point of like, where does, you know, study kind of evolve to like, where do people focus, like, you know, ten years ago, the the long list is unlimited computer science jobs. Go do that now. Maybe you might or some people might argue, well, the I will do all the programing in the future.
00;56;33;12 - 00;56;56;10
Unknown
I don't think that's quite the case, but that's kind of a, you know, the other direction. The natural question is where we'll be in five years will be more demand or less demand. I think there'll be probably more demand, as people say, oh, wow, so, so much capability here. I need to maybe not understand going back to where the semicolon goes, but I need to be able to think about product and how the things, you know, these systems tie together and how to build the next one.
00;56;56;12 - 00;57;21;11
Unknown
But I do think that the thing that I focus on when I talk to young people is about curiosity, because I think the like, it's so easy to be focused on the one task that you're doing to not think about how that connects to everything around you. And I think that that, that level of curiosity is what makes someone successful and what makes them move forward is thinking about kind of where their thing connects to the world around them.
00;57;21;14 - 00;57;45;08
Unknown
And if they're able to make that connection, they can, you know, have a bigger economic impact. And they're optimistic. Yeah, yeah. That's fascinating. I, I'm trying I'm projecting, my wife and I have arguments about, self-driving vehicles. Right. And in my mind it's like, well, I, I know kids are driven around by their cars as opposed the other way around.
00;57;45;08 - 00;58;22;24
Unknown
I don't, you know, I love my children. I don't think they're going to be great drivers. At 1516, I was already proven out right. Yeah, yeah, yeah. No I'm not not them. But like but but like the versus I right. Agreed agreed. But like people aren't people aren't convinced with data. You know they're convinced with emotion the emotional impact of that data and, and yeah I mean I, I, I the way I like the way you put it because it's like, if I can't connect, how much better these self-driving vehicles, not just today, but in ten years will be, versus, you know, and what the what the outlook on our day to
00;58;22;24 - 00;58;53;16
Unknown
day life will be in her mind. It'll always be like, you know, the machine, you know, like, beware of the machine. Right? The machines will make a mistake. 1 in 1,000,000 times versus you win 10,000, then the machine is dumb, right? Because of that mistake? Yeah. Yeah, I think that's that's probably going to continue forever. But I think, you know, I still believe that if you look at data, it's very clear that, you know, right now we're at a precipice where, self-driving cars are safer than human driven cars.
00;58;53;19 - 00;59;09;08
Unknown
Like, when things, you know, obviously be new to Houston. I'm like, I hate driving around in downtown Los. Like, you know, I'm a I'm a city boy who grew up on subways. And it's just like, okay, this is not very fun. And, I would much rather push a button, say, bring me to Galleria, bring me to the restaurant, or bring me to wherever.
00;59;09;10 - 00;59;28;01
Unknown
And, and right now, I'm. I'm finding myself, like, gripping the steering wheel, hoping I can make it so it's. Yeah. Yeah, I mean, I'm actually a better driver than that, but, like, you know, now you're somewhere. Yeah. I do wish we had. We had better subway systems here anyway, I, I I'm so excited for for what you're building.
00;59;28;03 - 00;59;54;26
Unknown
I really appreciate, the time you spent with me, I guess. One last question. I, like, do you find that you're trying to build a little bit more methodically now because, of, like, how fast things are moving? Or do you feel like you have to outpace or keep up with how fast things are progressing? I think the,
00;59;54;29 - 01;00;24;05
Unknown
Well, the interesting side effect of the speed at which we can develop now, means that you can do more product more quickly. And I don't say that as a pure positive. I think going back to, you know, what we and our team look at is we have kind of an idealist that we often, you know, look to, to say, how do we kind of enhance what we're doing, but also how do we ignore stuff too?
01;00;24;09 - 01;00;42;22
Unknown
And I think it's, because I think you do need a focus to say, who am I? What am I? And that's been actually a really challenging part of our development because as I mentioned, when we started dealio, we were focused on all deals and wasn't. So we said, well, let's start to think about more proactively about our go to market.
01;00;42;22 - 01;01;04;05
Unknown
We're like, we're a little bit confusing to everybody because we're not quite for them specifically. And so once we started being something for a certain group specifically, specifically, we were able to say, okay, now I can kind of scratch your edge versus, you know, a general idea. And then I'm going to wait for your question, but I think it's prefer it.
01;01;04;08 - 01;01;08;23
Unknown
Daniel, thank you so much. Man. This has been I really appreciate the time. Of course. My pleasure. You're awesome. Thanks.