A Billionaire's Guide To Going From $4/hour to $1 Billion Net Worth - Dharmesh Shah

23:03And there's no objective value because there's no comparables. Like no one really knows what the actual fair value is. Now flip that on his head, let's say I went to that same property. I'm like, this place is fantastic. I love it. Let me tell you the things I love about it. Let me tell you why. I'm like all in love with this thing that you're selling right now. And you go through that and it has to be authentic, has to be genuine. I'm not making this up. And then I put my price out there. It's like, you know, here's what it is. Now the seller has to say, okay, well, am I really going to find, like, I don't know what the price is.

23:35This person seems relatively clueful. He loves this place. Am I going to find someone else that loves it more than Darmesh loves it? Probably not. Like, he solely gets this thing or whatever, so maybe his price is probably what the fair price should be, right? Like, he seemed like a reasonable guy. So that's a, it takes people's guards out. And this only works, by the way, when you have one of those kind of rare, it's hard to find comps or whatever, where the price is not really objectively known. It's hard to kind of triangulate to like a quote unquote true price. Anyway, that's my power negotiating tip of the day.

24:05Okay, great. And so you told me something when we were talking before about this first company and you said, okay, so I did this company and actually maybe there's more in that first company, but you told me something great that I want to bring up, which is then you did a second company. And this was one that I don't even know if it's on your LinkedIn. You're like, this is my embarrassing company. It's my, things didn't go well. And Sam, the story he told was he was like, I did the first company when I just didn't know. I didn't know anything about anything. So I'm clueless, blah, blah, blah. We're just making up as we go.

24:33Super scrappy. And it kind of worked. It worked. Then he's like, I'm going to do the second one. But now I'm so much smarter. Now I actually know what I'm doing. I have some experience. I'm going to do things the right way this time. I'm going to go raise money. I'm going to hire better people. And I'm going to get a better off. I'm going to do these other things. I'm going to do it proper. And then I guess that that business, you can tell the story here, but it didn't work. But you said a great line. just because I was ignorant doesn't mean I was wrong. Meaning that first time I did that first business, I was ignorant.

25:03I didn't know anything, but it doesn't mean I was wrong. Can you unpack that idea? Let's say there was an objective truth function that says, okay, this was in this particular situation the right decision to make in whatever situation you're in right now. That objective truth function, if we had run against all the decisions I made in that first startup, I think my hit rate was actually pretty high. And the hit rate of lots of founders because our natural instincts in that situation actually turned out to be good instincts, like being resourceful, not spending too quickly or whatever, taking your time, doing all the things that kind of came as natural instinct.

25:34And I was not a natural born entrepreneur. It's probably likely why I have the insecurity. But in something like a startup or entrepreneurship, there's a bunch of decisions you're going to make. A lot of them maybe may or may not be optimal. But what was definitely suboptimal in hindsight was second guessing myself on the second one. It's like, I'm going to do the opposite of what I did in my first one because it's like, what a chump I was back then. So I'm going to go, I'm going to do a speed run. I'm going to do it faster. I'm going to write a$500 ,000 check to myself on day one to fund the thing or whatever.

26:05It's like, who has time? We don't have time to go through those kind of cycles or whatever. Really quick, I remember a time in high school, I had this like hillbilly friend that had like a huge chest because he was like so strong at bench pressing. And we went to like an exercise class and the teacher was like, the best way to get strong in your chest is to do incline bench press at a 30 degree angle. A 30 degree angle is so much better than a 45 degree and so much better than a 10 degree. And my redneck friend who didn't know anything, he just would go down and lift weights. He was like, I must have been doing them at 30 degrees because I'm strong as shit.

26:38It's like the same thing. It's like, dude, it doesn't matter if you know what you're doing or not. It's always the end result. What was the second company? I had no idea that you had a failure under your belt.

26:49Yeah. So it was a kind of Facebook before Facebook for small business. It was effectively a CRM. So I've been working in CRM now for a long, long time, built for small businesses. And this was right in the 2000 timeframe. So right as the bubble was bursting, and the first company was not an internet company, this was. It was a web-based information management tool for small businesses around customers. What was it called? It was called Captivo. C-A-P-T-I-V-O. How much did you raise? And how long did it last before you shuttered it? So, once again, because I was going to do things differently this time, I wrote the first$500 ,000 check.

27:26And I kept writing checks. I'm like, I have money. Why would I not just fund it myself? Can you say how much did you have? And how much of that did you put into the second one? Can you say that? I think I probably put in about$2 million, I'll say, give or take. I googled Captivo Darmash and I found a PDF that has a lot of handwriting on it. And it looks like it's a case study for Sloan. And you're trying to make a point. And you said, in 99, I founded my second startup called Captivo. This is what it was. It was in many aspects similar to Salesforce.com. And about two years in the product development and over a million dollars of capital invested, mostly of my own, Captivo could still not gain any significant traction.

28:07And ultimately, it was sold and it didn't work out. And so this is like the only thing that I can find about Captivo on the internet is a case study that you helped write. By the way, I mean, it's like we sort of take it for granted now. Part of it would be why, and it's not like, you know, I'm not trying to hide things, but the internet wasn't as big a deal back then. So yeah, I wasn't blogging, no one was blogging. So it was like, okay, well, why would you just say things? Like, where are you going to say them to? But so one of the kind of interesting things as far as kind of closing out that particular chapter, what ended up happening, I'm going to make my third sports ball reference, and this one's a golf one, it was a long put to par.

28:43What happened was I took that company Captivo, merged it into my first startup, which I had not sold yet. So I was doing two starts at the same time, which is not something one should do. And then ended up selling the kind of merged entity. So I ended up making my money back for all intents and purposes. I didn't really lose anything. But it was a very, very long budget par i just yeah um so the big idea so far were yeah uh at the beginning you're going to trade time for money and then at some point in your life you're going to desperately start trying to trade money back for time uh that's life i think that's that's a golden nugget the second one is cool you can either get more hours or increase the value per hour and that's where you started taking budget and investing it in books and training and seminar whatever whatever you can try to do to increase your own value you get the job at us steel that's where you realize you're either an asset or a liability to this company.

29:35You're either overhead or you're the one actually creating value. If you want to be higher leverage, you've got to be closer to where the actual value is created. So then you switch, you do a software company, you do that. The lesson around negotiation. So go back and ask for more and specifically not just ask for more, but what would it take for me to be making more? It's more like the question you asked your boss. Is that right? The actual lesson I drew from that is the power of being able to reduce and frame something in a very simple, like inarguable way, right? Like the fact that it was so punchy, right?

30:09Like it was just that one sentence, I don't need to find a bunch of companies. I just need to find one. And I think I can, right? Like, had that been like a five minute long conversation, me going back and forth or whatever and trying to make an argument, I don't think it would have worked. What made it work was the fact, and this is, I think, is a very, very, it's like anything else, it's a developable skill. And you are actually, Sean, are the master of this. is being able to what I call like insight compression, right? And I just made that up. It says, can you take some big idea and kind of boil it down, boil it down to like, it's the essence is still captured and it's just a dense distillation, for lack of a better term, of that idea of that concept down to something really simple because the simpler you can make it, the more likely it is to be transmissible, to be communicated, to be whatever.

30:55And this applies to so many, obviously applies to, you know, copywriting and marketing and things like that, applies to venture pitches, applies to so many things. By the way, I think one good point to make in this is this isn't just a mindset you had when you were young and broke. Like you took my power writing course a few years ago. It's like, dude, this guy's a, you're a founder of a 20,$30 billion public company. He's sitting here in the course learning and seeing, can I learn something about writing? Can I get a little bit better at this one skill? You were still doing that now, not just when you were, you know, 20 years old.

31:28It's like, oh, I'm a beginner. I got to like, you know, start to level up. It seemed like you continued doing it. A guy came on the podcast the other day, Mike Posner. He's a musical artist. He talked about he had had his first song go huge. So he thought, oh, that's what I do. I make songs. They blow up. That's me. I guess it's hard for these other people. It's not hard for me. So the first song is like, you know, whatever, five times platinum. His second song actually was still like double platinum, but felt like a huge failure. Third song, single platinum. He's like, I'm going downhill. And then the studio shelves them.

31:56And for years, they're like, it's too expensive for you to even produce the music. so we're just going to keep you, you know, you're on the shelf in the music industry. I was like, what were you doing during that time? And I thought he would just say he was depressed eating Cheetos on the couch. He's like, yeah, I was depressed eating Cheetos on the couch. And then I dusted off the Cheeto dust for a minute. He's like, and I enrolled in a Berkeley college for music. And he's like, I was learning to improve my singing. He's my ability to play instruments. I couldn't play the guitar before that. So I learned to play the guitar.

32:23And he's like, yeah, I just figured, okay, this period of my life, I'll just sharpen up my skill. I'll become a better artist. and he's like, I would go to these classes with a bunch of college kids and here I am, you know, Grammy award winning. I've been on the charts. I've made millions of dollars as an artist and there's better singers in here than me. And he's like, it was a real mindfuck. But what I took from that whole story was, damn, that's kind of inspiring. Like this guy used that time to sharpen his skills. I think a lot of adults just stop. I think they just, oh, that's stuff you do when you're young and you just don't need to invest in learning anymore after that.

32:57So talking about the power writing course that you have, and then just kind of writing and copywriting generally, is that the return on time for developing writing, I have not found the ceiling yet, is that it is just so high. I cannot describe to people, like if you could do nothing else, let's say you had five hours to invest in the next month or something like that, you could spend it all on learning how to write well. And future you will look back on those five hours and say, boy, that was a great use of five hours, right? Or 10 hours, whatever the number is. Because that's what I consider to be just an amplifier of things, right?

33:33It will make you a better thinker, better communicator, better pitcher, better salesperson, better everything. We were talking about being close to the action. I don't remember if you talked about this on the pod or if you talked about it just privately, but I had known that you bought chat.com. And we have known that you've been about agents. I think last podcast, we talked about vectors and things like that. I think we mentioned agents. And it seems like that's a big thing to you right now. And speaking of being close to the action, you bought this thing chat.com I think you said on the podcast it was something like 10 million dollars or maybe you said eight figures I forget and then it comes out like two weeks ago that you sold the domain chat.com to open ai for around eight figures and I think it said it was like all well we should add one thing when you came on the pod you had just bought it and we were like what are you gonna do with it you're like not super sure yeah I'm gonna try some things um but I just think it's a great domain and think it's a great investment.

34:35And you just, you had made a bet, but you didn't have the, it's not like you had it all figured out and it was all de-risked. Yep. You made that bet then. Now here's the update. What happened? Okay. So a couple of things just to kind of, so I think the time I went on the pod was like within like 72 hours of when that kind of purchase that happened. So it was just, it was just done. It was not distilled in my head. I actually had a plan for it. I was going to build a chat application on top of the GPT algorithm, similar to ChatGPT. And the thesis, and there have been two times where I accidentally competed with OpenAI, which I don't advise anyone to ever do.

35:13He's literally on the top of my list of people I'd ever want to compete with. That is Sam Altman. Just too smart and even a more red-blooded capitalist than I am, actually. And I mean that in the most positive way. So my original thought was, it's like, okay, they put this ChatGPT thing out there. It's a demo app, right? It's there to demonstrate the power of large language models and the underlying GPT algorithm, but OpenAI is a platform company. And there have been stories about these kinds of things, which is, oh, someone invents a cool new technology. I think PayPal started that way. They had some payment transfer thing, and they had an app that says, oh, here's our encryption technology demonstrated by me sending you money over a Palm Pilot or something like that.

35:50And they're like, oh, well, that encryption we don't care about. We care about the actual transfer of money part. Is it the story with OpenAI? like they were doing this thing to sell to software companies and then someone was like, dude, can you just make this thing so we can show people what we're working on? Yeah, and even Sam has kind of gone on record and said that. They had not expected it to kind of go. This is a way for us to kind of demonstrate the technology and kind of make it accessible so people can try it. And so it sort of caught them off guard as well, which is like this was not expected they were going to get tens of millions and it was 100 million users within a couple of months.

36:21But in the back of my head, I'm like, okay, well, that's great, but they're going to go back to being a platform company but someone should actually create something that is any user application that sits on top of LLM like GPT-3 at the time and then it just so happens the universe sort of configured itself and chat.com in a random event became available for sale for the first time in 30 years that had never been actually used before and I'm like, I need to do this It was like, that was kind of thing number one. It's like, I could use it for that. Thing number two, kind of rationalization to myself is that this is the cover charge to get into the AI party, that people will take me seriously.

37:04It's like, no one really knew me in that space because I'm not from that world. Didn't go to machine learning school at Stanford or something like that. But that's sort of something that's sort of hard to ignore. Like deep down inside, I'm part marketer, right? I think it's like, oh, this is a good story. It'll get people's attention. Did they reach out to you with the idea of acquiring it? Or were you at a conference with them and you're like, hey, I have this thing if you want it? I'll tell you. I'm going to draw the line in terms of at the transaction beyond because my experience, I think I'm at liberty to share what happened post that decision.

37:42I'm less comfortable sharing. The way it came to be is that I was at a Sequoia event. I think I actually talked about this on the pod. And one of the things that Sam announced at that event, Sam Altman was there, was this, oh, you know, ChatGPT, you guys know and love? Yeah, it's awesome. We're going to support these plugins to ChatGPT because right now the limitation of ChatGPT, it can't access third-party data source. They can't really do anything. And it's got the data that it's got based on its training data set, but it's like a snapshot in time. But these extensions will sort of amplify, multiply the capabilities of ChatGPT.

38:19and so that's sort of when the little switch went off in my head I'm like crap OpenAI wants to create the ChatGPT of ChatGPT they're going to turn this into an actual platform an actual end user app and so that's when I'm like okay, if I'm not going to do something I think it would be unwise for me to keep that domain I don't want to sit on it I don't have plans to do something with it I did not want to go into competition with OpenAI because it was clear that this was going to be a big bet for them and I was treating it as somewhat of a side hobby project. So I reached out to Sample. I knew. It's like, I don't know if you were interested or...

38:57And there were other bidders for the domain at the time that it went for sale. And my third back of my motivation was I had a suspicion of who some of the other players might be. I didn't want them to have it either. This is the kind of competitive side of me. Did you sell it for a profit or was it just like you... I think you said on Twitter you sold it for shares of OpenAI. Was it like, man, I'll just give it to you for the cost in order to be able to invest in your company? I thought I did this cleverly. When I made the announcement that I had sold it, I didn't share the details, but I provided a GPT prompt that says, if you type this into ChatGPT that says, oh, Dharmesh likes to buy domain names, but he usually does it because he's got a project in mind.

39:38Dharmesh doesn't sell domain names at a loss because he doesn't have to. Dharmesh also doesn't like profiting from his friends, and he considers, and he's known Sam for a while so OpenAI did buy ChatGP. These are the facts that you know. And then it's like, and then the prompt is if you had to guess, what do you think the domain sold for? Because I'd also disclosed I was a shareholder in OpenAI. This has happened a while ago. And so it's like... You weren't a shareholder before that. I was not. But you would have loved to be an investor in OpenAI. And this was, like you said, the cover charge.

40:11It was the ante in order to get there. This was like even a even a better party, right? This is the shareholder party. I think based on all that, I think I put your prompt in and it was like, Darmesh bought the domain for$15.5 million and now owns $15.5 million of OpenAI stock. That was the AI answer to it. You can leave it at that. Yeah, we'll leave it at that. That's so funny. Do you have any? One thing I have to share with you, and this is the non-humble part of it, but there's a lesson in here. So there's an old Steve Jobs quote around connecting dots. And the quote goes something like, you can't connect the dots looking forward.

40:52You can only connect them looking backwards. So you have to have faith. Faith in something that the dots will connect at some point in your future. You just have to trust your instincts. And I've been a big believer in that even before Jobs said those words, which is you sort of collect what he calls dots. And I really like the idea because I love graphs. Those are things that I love. It's like, okay, well, if I made an inventory of all the dots that I've collected through my life, and dots can be people I've met, can be skills I've learned, things, experience, lessons. And then over time, if I had to go back and write a history of my life and say, oh, well, this was kind of a weird dot at the time.

41:28I could never have guessed that had it been not for that dot, this over here probably would never have happened, right? You don't know in a parallel universe what was actually going to take. But I can tell you for a fact that the universe as it exists right now would not have happened had that particular dot not been collected, that I had not collected it. And so my big lesson here is that you have to spend some amount of your time, and this is the same kind of lesson, just phrased differently, around investing in those dots that may or may not make sense at the time that you're doing them, but they feel right.

41:58They feel like they could have something. It doesn't have to all be kind of maniacally diabolical and capitalist. Sometimes it's just like, I want to collect this dot because I love this. I believe it. I have conviction. Whatever it is, right? It's like, it's you. it's your time on the planet, it's your money, whatever it is you're doing, you should be able to kind of collect those dots. But if you kind of trust your instincts, and not all the dots are going to work out, and this is the beauty of these kinds of things, is that you don't need for all of them to work. You just need one or two to work really well.

42:25And that's it. That's literally it, right?

42:39That is the banking product I use for not one, not two, but actually eight of my businesses. I have eight Mercury accounts. I just went and counted. I use it for every one of my companies. It's an absolute no-brainer. Over 200 ,000 other fast-growing, ambitious companies use Mercury. It's one place you can go where not only you can, of course, have your money there, but you can send invoices. You can pay bills. You can create reimbursements for your team. Pretty much all of your financial needs can be housed under Mercury. And the product is beautiful to use. And the reason why is because it's a product that was not made by finance people.

43:09It was made by a founder, Ahmad. He's been on this podcast before. And he used tons of products as a startup founder. And this is the one that he wished he had. And I actually reached out to them to become a sponsor for this show because I'm such a big fan of it. So if you need a banking product for your startup, use Mercury. You will not regret it. It's amazing. For more information, check out mercury.com. Mercury is a financial technology company, not a bank. Check show notes for details.

43:36I'll use a sports reference. in basketball, there's people who you say have a nose for the ball. They just know where the ball is going to bounce. They know where to be. The ball just sort of always ends up in their hand, end up in the right spot. You talked about your first job application was to Pizza Hut, back when you're just moved to the country type of thing. But pretty quickly, you found yourself in software development in the 90s, which was arguably the best place to be. Then you said in 2000, you started a web internet company. Internet was the next best place to be after that. Then you create HubSpot, which was a cloud SaaS company, which is arguably the next best place to be for that next decade.

44:12And now you just sold chat.com to OpenAI. You own a bunch of OpenAI and you were talking about AI agents. And I would say that seems like in this decade, the best place to be. And so I view you as somebody who has a nose for the ball. and what I mean by that is like the frontier that you're interested in seems to be the frontier where all of the action and the value and the trillion dollar companies are going to get created which means if the trillion dollar companies are created that means the tens of billions of dollar companies are going to get created, the billion dollar, the hundred million and the ten million dollar companies also, it just means it's the target rich environment so I think that when you say this is the thing I'm most interested in we should pay attention tell us why agents are the thing we should pay attention to and kind of bringing it back to the kind of early, just as a thank you for the lead-in, that was great, is that we talk about you, the audience, making your first million or your next million if you're further along in the journey, is that that first one is almost always harder than subsequent ones, and that's not counterintuitive at all.

45:16I will say this, though, is that never in my mind has it been easier to get to that first million than it's going to be here, as we're living our lives right now. And part of that is what agents unlock. So let's talk about agents. We'll talk about what they are, why such a massive opportunity, and what you, the audience, should do about it. Okay, so we'll take a step back to last year. Last year was all about chat, and everyone's used chat GPT. We all get it. We type in a prompt. Something comes back. Awesome. And that's what's a very kind of interactive approach to the use of AI, right? Like you type a blog.

45:52It's like, oh, give me a blog post or do this for me, and it comes back with a response, and you may do a follow-up. But it's effectively this kind of back and forth interaction model with AI. What agents are, and as a result of that kind of interactive model, the tasks that you assign AI are generally more discreet. That says, okay, do this for me. There's a single artifact, produce a blog post, generate image, whatever it happens to be. What agents are, kind of very simply, are it's AI software that can accomplish higher level goals requiring multiple tasks, I mean multiple steps. So it's not just a one shot, give me a blog post, and it comes back with a blog post.

46:29It's like, I want you to do this thing, and that thing may require 10 steps. And each of those things, as we were just talking about earlier, it might need to functionally decompose everything in order to accomplish that higher goal. And it has to have memory, it has to do all these things in order to kind of make that possible. But that's what an AI agent is. And right now, we're going to talk about agents as it currently stands, my expectation is that agents are the new apps. It's just software, right? So when mobile came along, it's like, oh, there's an app for that. There's an app for that.

47:00Not that far in the distant future, everything is going to be, there's going to be hundreds of thousands of millions of agents, right? The same way that there's lots and lots of apps out there. So think about agent, if it's simpler to do that, it's like, it's just an AI app that happens to do really high order or can do high order goals requiring multiple steps. So the thing I think, without getting into agent.ai too much, here's kind of my view of the world in terms of how this is going to shape out. There's lots of debate going around, oh, in order for it to be an AI, a true AI agent, it has to be autonomous.

47:32You can't have, you know, it's like you just have to be able to give it a goal. I'm more pragmatic than that. I don't think that's a requirement in order for something to be called an agent. But here's the important part. is that, so if you kind of pull out that thread, the way the world looks soon is that we will have hybrid teams that consist of humans and consist of AI agents. And the easiest, simplest, and therefore in my, I'm going to say this with some conviction because it's better to do it that way. The easiest way to pull that off and the simplest way to describe it is if you think of the digital agent as a digital team member, because that's the easiest way to get there from here, right?

48:14It's like, okay, if you just say, I'm just going to hand it over and the agent's going to create the, like just do all the things, there's going to be no humans. Yeah, that might happen someday, but it's probably not going to happen as a direct jump from where we are today. What's likely going to happen is individual tasks will get automated and we will start injecting and having these hybrid teams. In the same way we had hybrid teams before we had like all full-time employees. Like, oh, I'm going to hire a freelancer for three months to do this because we don't need them forever, for whatever reason.

48:39You did that. And they're like, that's okay. They don't work in the office. Then we had a hybrid from a geographic perspective. And now we're going to have hybrid from a carbon-based life form versus non-carbon-based life form perspective. It's basically like this. It's like, we've got guests coming over. And now my wife decomposes that into, okay, we need to clean the kitchen. We need to clean this area and we need to get food. We need to order food. And so she tells me, load the dishwasher. I go load the dishwasher. Then I hand it over to my agent, my dishwasher agent, who then will wash those dishes.

49:09And so I'm kind of the, she decomposes it. I'm a human in the loop. I'm doing one step that hopefully, ideally, that we could have another robot that loads the dishwasher. But for now, I do that. But the dishwasher is better at doing the dishes than I even was. And we'll do it well every single time. I trust that agent to do it. That's what the inside of companies is going to look like. Yes. And it might be that, you know, I think we will have humans doing the review and approval for a vast majority of tasks, right? So I think the digital team members, for lack of a better term, are going to be doing lower level tasks to start with, because that's the things we can trust them with.

49:44It's like, okay, well, the stakes are much, much lower. It's like, okay, produce 10 versions of this blog article. It's really super important for my business. But then let me pick the one that's actually going to go out. And then once I pick the one that's going to go out, because I'm a step in the process, the human doing that, then the post-production maybe is all digital. We don't know, right? Give us an example of an agent you're using that you actually use now, because on a day-to-day basis or week-to-week basis is actually live. It's working. It's not just a cool demo. A nice simple example is I have an agent that says I really like this episode of, let's say My First Million.

50:17I recall hearing something about X. Some word stuck into my mind. I want to write a LinkedIn post based on that. Here's what I want to happen. All I'm going to give it as input, this is the instructions kind of MPC model. It's like, okay, here's what you get from me. is a YouTube video. What I want as an output is a LinkedIn post that's going to do well by some definition of do well. We have training data on that. Okay. What it's going to do is going to say, okay, first step one, pull the transcript from a YouTube video. Step two, figure out who the players are. Step three, highlight the things that are quote worthy, tweet worthy, remark worthy, whatever it was happens to be.

50:55Step four, figure out what the hell Darbesh was actually asking about. There was a snippet in there that he made reference to. Clip that as well. Put that all into the thing. Now go say, okay, like, what style happens to work well on LinkedIn specifically? I was like, write me a 200-word prompt that describes the language and the style, whether it's bullet points, whether it uses emojis, doesn't use them, like what works. Have an agent that does that, like a style creator thing, whatever. Take all of that and now produce a LinkedIn post and just give it to me. And so you do that? You have a thing that does that?

51:27I have a thing that does that. And by the way, the really fun part about this, it's like Lego bricks, right? So I have an agent that does a YouTube transcript. Bothly straightforward. But it's a better YouTube transcript than the one that you would just get copying off the web because there's an LLM involved. It's like, I want a YouTube transcript that has chapter headings, which is not going to be in the transcript. That's going to bold things that are actual quotes or whatever. It's going to format it in a way that's human consumable. Way better transcript than the regular transcript you get out of YouTube.

51:53So like, check this out, Darmesh. Have you seen this? Go to mfmvault.com. mfmvault.com. My friend Greg is building this. and it's not ready for the big time, but here we are. Let's do it. So mfmvault.com, you land here. So he built basically a site that kind of works like what you described just for specifically My First Million podcasts and fans. Okay, so he's like, all right. He was like, I don't just want summaries. I want, he's like, you know, I like MFM because it's got ideas, it's got frameworks, it's got stories. And those are like the big things that he really cared about. So if I just go, like, first he's got just like a bunch of, and AI is basically extracting those from every episode.

52:34It's like searching for frameworks, searching for stories, et cetera. But like, let me just show you this. If I go to episodes, we'll try this live, see if this works. And I typed Darmesh and this is our last episode we did in 2023. So it's got the, you know, the overall summary with chapter titles, like you said here, but check this out. So if I go to stories, it's like chat spots,$1 sale, Pandora's Manual Musicians. It's like the 17-year evolution of chat. And if I click that, it'll take me to that moment, but it'll also summarize it. There's also frameworks in here, like an AI immersion week.

53:07Basically, the idea here was how I dedicated a full week to just hands-on experimentation with AI tools only. And so it is extracting and categorizing and linking these ideas from the podcast this way. But to do it, like you said, he's strung together five agents. There's the listen for when there's a new episode. Then that guy says, I found a new episode. Pass it to the transcriber. The transcriber's like, I transcribed it. Pass it to the summarizer. And the extractor passes it to those guys. He's like, great. Turn that into clips. Pass it to the clip guy, right? And each one of those is their own agent.

53:40And I think that's, and then like you said, it's composable Lego blocks. So if I want to use a summarizer for something else, well, once that agent exists, he can summarize anything, not just MFM episodes. Yep. That is the future, Sean. So a quick one-sentence description, this is the positioning, I'm going to test it on you guys as well, is agent.ai is the number one professional network for AI agents. It is also the only professional network for AI agents. It's LinkedIn for agents? Yes. And so the question is, why do agents need their own professional network? That's a reasonable question to ask.

54:19And the answer is, if we imagine that I happen to be right, that we're going to have this hybrid world where we have both AI teammates and carbon-based life form teammates. Well, how are we going to find those people, right? So we have Fiverr for humans that we can sell. I need a logo created for whatever. I can go to Upwork. I want to hire someone for the next four months to do whatever skill set or whatever. There needs to be an equivalent of that for AI agents, right? So I don't know what's out there. Like, I don't know. And here's where it starts to get really kind of super cool. So imagine a professional network and it has ratings and reviews.

54:52It's got their experience. Like, oh, I'm an agent that does this. I've been used 40 ,000 times. And here are the people that like me. Dharmesh follows me on agent.ai, right? Because humans can follow agents. Agents can follow other agents. And what's going to happen over time is that agents will be able to hire other agents on the network because everyone knows what everyone else can do. They can try other agents out. It's like, oh, I've been given a budget of$100 to try five different agents to figure out which one is the best for my particular use case. and it can go without human intervention, just go try that.

55:23It's like, oh, network, yeah, I'll pay you. This other agent can do this and this. So over 30 years, I built a lot of software that is what I call solo software, right? Like just like software for me, for my public speaking, Conti LPMs in a talk, that kind of stuff. And then I found that most of those things can be better with AI. Did you hear that little throw in LPMs? Laugh for minutes. I like it. Yeah. By the way, I did not invent that term. Stand-up comedians use that. That's been around forever. um so yeah uh i stole that from sam it's something that funny people like us we know we wouldn't expect you to understand that jermich can i ask you a um i'm gonna ask you a rude question and then i'm gonna ask you a thoughtful question but i'm gonna start with the rude question rude question is this dude you're a billionaire and you're super smart and you love all this new tech like why aren't you doing the elon thing elon's like cool i did my zip too i then I did PayPal, then I did Tesla, SpaceX.

56:19I'm going to keep kind of building new companies, bigger, badder bets. You've been at HubSpot for so long. You've been doing CRM for like 30 years. Don't you feel like you want to just spread those wings, baby, and just fly and just go build your rockets? And why aren't you doing like a next grand act? That's my rude question for you. Yeah, no, it's not rude at all. And the answer is actually quite simple and has the added value of being true is that this sort of is my next big act. So the problem is, okay, so let's say you're playing a video game. Pick up your video game of choice. And you've kind of ground it out.

56:57You've built the things. You've got the weapons. You're there. Like you're now in kind of power mode you can do. Now, let's say it's Zelda or something like that, which I love, one of my favorite games. And now you can just go out exploring and try things and do stuff because you've got all of that. If you're going to play a new game, there's a bunch of stuff you still have, even though, because you have to sort of get right, you're going to have to build a team. I'd have to go find a co-founder. By the way, I've won the co-founder lottery, right? Like I didn't, like no one knew it at the time, but that has a massive impact on your outcome.

57:26It's like, there's so many things I'd have to get right just to be able to kind of do the next thing. I think my odds are doing the next, like even something like agent.ai, my odds are higher pulling that off with the infrastructure I already have beneath me in the form of HubSpot. And sometimes I have to squint a little in order to kind of squeeze the thing and make it fit in HubSpot-shaped box, and I have mechanisms for which I can kind of do that. But I don't feel like I've ever been kept from doing the thing that would have been my big, bold idea as a result of being a CTO at HubSpot. If that were the case, yes, I would go, you know what, it's been a great ride.

58:02Wish the company well, and we'll continue to be a cheerleader. I don't accept your answer. Why don't you accept it? Well, let's break it down. He's a logical guy. Let's break it down. So let's say he has things he's excited about, new technologies, new ideas, just potential, potential to create cool shit in the world. All right. So there's two ways of looking at this. Either HubSpot, HubSpot is the best place to do that, which I would just view like, I'm not saying the infrastructure, I'm just saying like the company HubSpot, which has like a specific mission and set of customers it needs to serve and like existing business model and all those things.

58:41Like the odds that your creative brain would latch onto an idea that also was like, should be on HubSpot's roadmap just seems like unlikely to me or just like an unnecessary limit that I wouldn't put on. Okay, so then let's say there is an idea that you get excited about, your version of SpaceX or whatever it is, your Neuralink, whatever the idea is. Either, well, HubSpot's not limiting me and I'm getting these benefits. Okay, cool, there's that trade. So it's kind of like, what would you gain if you actually were out on your own? You would like, there's like all these things, like when people have their job and then they try to have a side hustle.

59:13There's one great thing when you go and you quit your job to do a startup, which is that you've now quit your job to do a startup. You've now told the world, you've told yourself, you've told everybody that like, I'm going to make a thing because I just gave up a thing. And that means I got to have a new thing now. And I think that the burn the boats thing is like probably one of the only assets or advantages a startup in general has or an entrepreneur has is that they just go all in on something. They don't have any meetings about anything else. they don't have any distractions and they've told the world I'm going to do this and I think there's something very powerful about that and I think the if I really wanted to turn the knob and be like what would be the and by the way you don't have to do this I'm not saying one needs to do this in life but it's like let me see what I can let me see how fast this car can go let me see what I'm capable of I just view that you're more likely to see the highest potential version of you outside of HubSpot versus doing side projects while using HubSpot and then doing talking at inbound and then taking these meetings.

1:00:08There's got to be the unblended version. The concentrated version had to be more potent than the blended version. So that's my case. Sean's trying to convince a billionaire of a founder of a$35 billion company. Like, hey, what are you going to... It's the ant telling the elephant how he should be walking, right? What are you going to quit your job and actually do something with your life and go in on your side project? Take it as a compliment. It means I view you as an Elon-level character, you know? That's very high praise. Thank you. I'll say this. I think it needs to be said. I am a big, big believer in kind of taking that leap of faith, right, and going out and doing the thing, and that's great.

1:00:50And I'll say it's going to sound offensive, as I don't mean it to be, but as I kind of write the next chapters of my life as I'm looking out, one of the questions I've asked myself is like, okay, what does success look like? right like okay so like what do i want to reflect back on like of the things that i did right it's um and a couple things have kind of emerged from that reflection is i like to build things and like that is like the truest thing that there is i'd like to build things right um and so one thing i've kind of managed to do which doesn't happen often so i don't have any direct reports because that would keep me from building things right like if if ever there was a time And we're just like, oh, I can no longer build things as a result of X.

1:01:33Whatever X happens to be, I will break down that wall, whatever it is, Upspot or otherwise. So that's one thing. And I will say this. I think your point's really well taken, which is most of the time, we have these kind of externally imposed rules on us. For instance, if you're an engineer, it's like, well, you're a CTO. You shouldn't be writing code anymore. And I do not accept that thesis because software is a creative discipline. It's a very hits-driven business. so you have to be right a very few number of times. It's like, we don't tell a musician that's really, really good, oh, you can just stop writing music and doing music and go hire somebody.

1:02:07We don't tell exceptional writers to stop writing and manage a team of writers. Keep writing. Do your best to minimize the stuff that's not writing time. And that's what I feel like I'm doing. I'm maximizing my build time. And to me, my personal mission statement is to help millions grow better. That's the simple, I want to be able to look back on it and say I had a positive impact on the most number of people, like that multiply house, like X amount of impact, whatever that impact is on Y number of people. That's my truth function. And this is why I don't do one-on-one meeting because it doesn't scale, right?

1:02:41This is one of the reasons I'm here and not elsewhere. It's not just because I like you guys, it's because I can have an impact. And the reason I chose the topic, it's like, oh, this is what I want to talk about. It's like, oh, you know, what's the best way for me to use this time that if I look back on it or the people that are spending an hour and a half-ish, listening to this, was that a good use of my time? Did Darmesh deliver on the promise of at least marginally increasing my probability of making my first or next million? You do this a lot. You are telling a story and you have these little bits that are wildly fascinating.

1:03:14One of them being that you have a personal mission. Do you have personal values like a company does that you and your family or just you operate against? We've talked about it. We haven't done it. So the HubSpot values are effectively the founders' values. That's where the original company values came from. And most companies are actually a reflection of their founders. What is the most controversial or value that has the biggest trade-off? Not everybody would choose this, but we choose this. What is that for you guys? Not like integrity or whatever, things like that. Yeah, one of the ones that almost didn't make the cut.

1:03:51not because we don't agree with the value, but there were others, is humility, like humbleness, right? And like the common argument that comes back is like, we're supposed to be a winning team, we're aggressive, which we are, right? And so people often confuse humility with a lack of confidence or a lack of aggression, assertiveness, or all these things. That's not it at all, right? The humility is being self-aware, is being able to recognize and not being a know-it-all. It's like, I'm here to learn, having that kind of, and that is a large part of what drives HubSpot. And so I've fought for that value for a long, long time.

1:04:25The second most controversial one is empathy, which is our second value. And it didn't always used to be empathy. We changed it along the way. Sam, what's our values on the pod? Yeah, that's a good question. The reason I've been asking this is like, well, I've been thinking about do I have values? What are my values? It's kind of cool. Is denim a value or a jacket? It's kind of cool to codify it. But no, we don't. MFM, well, we kind of have values. We've had people in the pod who have done bad stuff and they're like, are you going to crucify me? And we're like, well, our default is not to criticize.

1:05:00It's to build people up. But sometimes that might happen. But anyway, so we have a couple, but it's interesting to kind of explicitly say these are the values of X, Y, and Z. And Mark Zuckerberg, I read this cool thing about him, or maybe he even told me in the pod, Sean, where he was like, a value means you have to sacrifice something. So move fast and break things. means we are going to move fast and because of that, I am okay with breaking stuff. Too many people, their values are things that are just obvious that don't involve a sacrifice. That are just platitudes, yeah. And by the way, there's not many of these.

1:05:36So for example, Move Fast and Break Things I think is in the pantheon, the hall of fame of values. Another one is Ray Dalio's Bridgewater thing. They have radical honesty or radical candor which is like, we are going to uncomfortably be honest with each other. And we just think that that maximizes in the long run, but in the short run, this is not going to be like anywhere else you've worked because we take that seriously. We take the trade-off, which is discomfort for the upside of being radically honest. I've just jumped to my end, and I think it's a useful thing. So we talk about maximizing, right?

1:06:12And I love that word. And many of you have likely heard of this notion of finding the maxim. and then the trap that people fall into is finding the local maximum versus the global maximum. And just from a layperson's perspective, it's like, imagine that you're kind of looking at a graph, like a normal, imagine a bell curve for those listening. It's like, oh, there's a point at which that value is the highest, right? It's the top of the hill, so to speak, in mathematical terms. The kind of local versus global max is like, okay, well, that's the highest point, but let's say you were to zoom out on that same graph and then there's this other bell curve right next to it, which is an even higher maxima.

1:06:45It's like, okay, well, if you were just solving based on that chart that you were looking at because you were zoomed in, you didn't actually find a maximum. You found the local maximum, not the global maximum. That's the, and so here's the, and so that one, that part of it is obvious. The non-obvious part is that often in order to even see the global maximum, you actually have to climb the local maximum hill. It can't be the abstract sit down the sidelines. It's like, oh, I'm going to go wander around. What you have to do sometimes is you have to make the effort to climb the smaller hill. You get to the top of the spot.

1:07:16It's like, oh, oh, now I see the landscape and there's that massive mountain over there, which is the thing that I'm actually meant to go do, right? And so part of it, and this goes back to the kind of connecting dots thesis as well, is like sometimes you have to do the thing that does not seem like the big bet in order to have the perspective to see what the big bet's going to be, right? Sam, have you heard this concept of RAS? Like instead of SaaS companies, RAS companies know what this is. I saw him put that on there and I was like, is that like a different word like Riz? What is... I was like, is this like RAS?

1:07:49Is RAS meaning like you're going to make fun of me? Like I'm going to RAS you? No, this is a big idea. Darmesh, can you do the RAS bit? So we all know SAS, which is software as a service. And we contrast software as a service to software as it used to exist back in my day. I can say that now. And back in my day, software was shipped in boxes on CDs and things like that. And you kind of plugged it into your computer and loaded the software up to do whatever you wanted to do. And software as a service was, oh, you don't actually buy the box, you don't buy the CD or whatever, we just provide the value of the software to you as a service over the internet, over the cloud.

1:08:26Results as a service is like, actually, you don't need access to software, you tell us what it is that you actually want to do, what's the outcome that you're looking for, and we'll just sell you that. You want the LinkedIn blog post at the end. A good example would be, He's like, oh, I could sell you legal software to analyze a contract in the real estate space that will give you commentary on whether this matches benchmarks or even in the VC world, a term sheet or something like that. It's like, oh, here's software that will help you power write a real estate contract and do the editing work.

1:08:56It's like HubSpot. Here's software that you can write blog posts, capture emails, and get more customers. Yes, but this thing is like, okay, well, what if you could just skip the intermediate steps? It's like, oh, you run more customers? What are you willing to pay for customers? And this idea is almost as old as time, right? Which is, because we've had like, in that particular example, we've had lead gen companies forever, right? Which is effectively results as a service. Like, don't worry about phone numbers. Don't worry about this thing called the internet or whatever. Pay us X amount of dollars every time your phone rings.

1:09:25Well, and it's also a sales technique, which is like, do you want guests to feel great at your home and to be popular? Okay, buy our vacuum to keep your carpet clean. Yes, that's exactly. Yeah, we've had that. Yeah, the copy is like, people don't buy, don't want drills, they want holes, right? The drill is a tool. Hole is the actual outcome desired. So RAS is the next wave of software. And there will still be software being used by whoever's providing the results of the service. That's how the thing gets accomplished. But the way it might be packaged and sold might end up being RAS. One thing, just closing out on the agent front as far as the call to action.

1:10:04So I forget the name of the guy that built the MFM vault. Greg, yeah. So what Greg did in terms of sequencing these things together, so he effectively built an agent. It has multi-steps. Each individual step may call other AI tools. Agent.ai not only is a professional network for distributing agents and finding agents, discovering agents, reviewing and rating agents, and talking about agents, it also has what's called an agent builder, which is a platform for building agents without writing code. You've had a good quarter. HubSpot stock, I think, almost hit an all-time high. You sold the business, or you sold the domain to OpenAI.

1:10:45Came on the podcast. You came on the pod. You know, I don't care what they say about you, Dermesh. You're doing okay. I'm doing okay. I'm doing okay. Thank you. I don't care what they say. You're all right. What I love about you is you have a bunch of things that sound like paradoxes. Like, you're extremely gentle and kind. You're also, like I would say, competitive and aggressive. and in other ways, I think you are humble, but then you're also a marketer and you're like, oh, I can see the marketing value of this and the practical value of this. But I also, you know, I'd rather be, you know, quiet, low key.

1:11:19I don't need to brag. And so you have all these like contradictory things, which I think makes you interesting. Hey, by the way, is this, you said there's, you go, as soon as I get on a leaderboard, I want to be number one. Yeah, you're like, but then I don't want to be on the list. Is this a leaderboard where you're like, I would like to be richer because it would be like, I think you might try to be cool and be like, no, I'm happy. But I think with every level you are on, whether you are with 1 billion, you want to get 10. Whether you have 10, you want to get 50. Just like, no matter how ripped you get, you want to get a little bit more ripped.

1:11:52How much of a motivator is like, you know, it says 1 on Forbes now. How do I make it say 10? 10 billion. Yeah, yeah, yeah. Well, it's a really good question. I tend not to have explicit goals. Goal setting is not something, and I'm not discounting it, I'm not disparaging it at all. Some of the smartest people have goals, they have those kind of really concrete things that they're shooting for. It's not that I don't do it now, I never really did that. It's like, okay, I want to get better, I want to have money, I want to be able to have freedom, make choices. I want to have the things, as I've said this before, I like to configure the universe to my liking, to the best of my ability.

1:12:33And whatever resources I can kind of accrue to help make that possible, I will go try to find a way to make that possible, right? But like, so like deep down inside my soul, I'm a math and physics guy that's not smart enough to be a math and physics guy, right? And my truth function is like I said before, it's like, okay, I'm trying to create, based on that mission statement, positive impact for the most number of people, like multiply X by Y. And the marketing is part of that. It's like, okay, well, I'm not going to be able to impact people if I'm not doing some marketing. It's like I can say the most brilliant things in the world if only 17 people look at them, then the value of X is 17 for that particular unit of work.

1:13:12And that's just not how I operate. I will make sacrifices and other funds in order to solve the truth, like truth function to solve for the mission for me personally. That was the balance there. We talked about a bunch of stuff. You had opening remarks. Do you have a closing remark? is there a way you want to end it? Did you achieve your mission of saying what you thought would help somebody increase their odds? My ask, my favor is leave a comment and let me know what you thought. That's where I keep scoring. That's the leaderboard I'm after right now. I'm number five on the YouTube list. I know comments and engagement actually help the algorithm probably more than lights do, by the way, for the record.

1:13:53Dude, you're the man. We appreciate you doing this. I have a notebook here where I take notes. I'm not going to turn it. I don't want people to see it, but I have filled three pages of notes. So you're the man. It's all a good pleasure to be on. Thanks for having me. Thanks for indulging my quirks. Dude, you're great, man. You're fun to hang out with. You are fun to hang out with. You're fun to talk to. I feel energized when I talk to you. It's good to see you guys. We appreciate you. That's the pod.

1:14:25it like no days off on the road let's travel never looking back all right this episode is brought to you by mercury they are the finance platform of choice for over 200 000 companies shouldn't be surprised because i use it myself for not one not two but i have eight different mercury accounts i have seven for uh different companies that i'm a part of and then i have my own personal account because now they have personal banking which is a really cool feature. I highly, highly recommend it. Like I said, I use it myself. And the reason why is because the way that Mercury works is beautiful. It's very intuitive.

1:15:00And you could tell that it's actually made by a startup founder. It's an entrepreneur. You could tell it's made by somebody who used other banking products in the past and didn't like all the different rough edges and annoyances and decided to actually fix it himself. And really any type of entrepreneur you are, let's say you're an agency. Well, one of the things every agency has to do is be able to send invoices, easily create them, send them to customers and stay current on your balances with all your customers. Well, you can do that inside Mercury. And so I think that Mercury is great. Highly recommend you check it out.

1:15:26And thank you for sponsoring the show. For more information, check out mercury.com. Mercury is a financial technology company, not a bank. Check show notes for details.