AI transcript
0:00:06 Hey, welcome to the Next Wave Podcast. I’m Matt Wolfe. I’m here with Nathan Lanz, and
0:00:12 today we’re talking vibe coding. In this episode, I’m going to break down exactly how I’ve been
0:00:17 rebuilding the Future Tools website completely with AI and how you can build your own website.
0:00:21 We’re going to break down a whole new business model that’s emerging because of these AI tools
0:00:28 and show you a business that’s doing $1.4 million a year from one single creator. I’m going to
0:00:34 break down what MCPs are, what APIs are, what all of the acronyms are, and we’re going to deep dive
0:00:42 into how you can build your own businesses and products and different tools using this AI vibe
0:00:47 coding concept. So let’s go ahead and jump right in. Today, we’re going to talk about vibe coding,
0:00:55 one of the hottest buzzwords in the AI world, and it’s all about vibing and coding, I guess.
0:00:58 How would you define vibe coding, Nathan?
0:01:03 Actually, I know Andre Karpathy. He’s the one who coined it. I think of actually Riley Brown,
0:01:07 because he was the first person I saw talking about vibe coding, really. He was sharing all
0:01:11 this stuff online. We’re like, hey, I’m just chatting with Whisperflow, and then it’s transcribing
0:01:17 what I say into text, and I feed that into cursor, and then it makes something. And he’s kind of
0:01:21 like describing that whole process, how different that is versus you sit down and you got to code
0:01:25 everything by hand. You got to really think it through. It’s a different process. So when I think
0:01:29 of vibe coding, I think, you know, in theory, it’s a more relaxed process, right? You could even,
0:01:34 in theory, be doing something else. You know, like in the evenings now, I find that when my wife is like
0:01:38 taking her shower and getting other stuff ready and other things, I’ll be like hanging out on the couch,
0:01:41 maybe watching something on YouTube, educational and vibe coding.
0:01:46 Dude, I do the same thing. Yeah. So this is like become a common thing for me. We’re like now in
0:01:50 time periods where I probably wouldn’t have been productive in the past. I’m actually productive
0:01:54 during those time periods. Like I’m like watching some educational and making something. And it’s
0:01:58 actually kind of fun to make stuff now because you can just chat with your computer and it does the
0:02:03 thing. Yeah. I mean, I usually do it in my office. I got a sort of double monitor set up here,
0:02:08 right? So typically what I’m doing is on my main monitor right in front of me. I’ve got,
0:02:12 you know, my coding app open my IDE. I’m usually either using cursor or windsurf. I sort of switch
0:02:17 back and forth. We’ll talk about that in a few minutes here, but I’m got my coding going here
0:02:23 and I’ve got YouTube typically opened up on my right screen and I’ll give it a prompt of what I
0:02:27 want it to go code for me and then hit enter. And then I’ll press play on my YouTube video,
0:02:32 sort of watch for a minute. Notice out of the corner of my eye that the coding has sort of finished
0:02:36 pause my YouTube video. All right, let’s check out what you did. Right. All right. It’s not quite
0:02:42 working right. Go fix this. Enter press play on my YouTube video again, go back to watching it.
0:02:47 And so it’s almost like you can code passively, right? Or, you know, I’ll do the same thing.
0:02:51 I might hang out in my living room, grab my laptop and just be sitting there on my laptop doing some
0:02:57 coding while the TV’s going. But yeah, I mean, essentially I think vibe coding is using natural
0:03:02 language to write code. I actually looked it up on perplexity to see how perplexity would define it
0:03:08 based on its research. And it says coined by Andre Carpathy in February, 2025. This method involves
0:03:13 describing software ideas in natural language, allowing AI to write the code and iteratively
0:03:20 refining it through conversation. So basically just talking to a coding app saying, here’s what I want
0:03:25 you to create. Go do it. Letting it go and write the code. You double check it and either saying, cool,
0:03:30 that worked or no, that didn’t work, go fix it. And just going back and forth using natural language
0:03:35 until eventually you get the app that you wanted to get out of it.
0:03:40 Right. And there’s been a whole conversation on X around this too. Like engineers are really skeptical
0:03:44 of vibe coding and like turning it into a meme, like in a negative way. They’re talking about like vibe
0:03:48 coding, you know, the systems for like how planes are flown. Like probably don’t want that, you know?
0:03:54 Yeah, I agree. Right. I think there’s definitely critical like infrastructure areas where you don’t
0:03:59 want to hear that it was vibe coded. Right. Yeah. I don’t want the software that the air traffic
0:04:03 controllers use to be vibe coded. Right. Right. Right. And a lot of times right now, the AI, when it does
0:04:07 make things, there are typically like small bugs and there’s probably things that are not the way that
0:04:11 you would have hand coded if you were like an expert. I kind of believe over time, those things are going to
0:04:15 just get fixed. Like the models are going to get better. You know, it’s definitely a big debate on X right now.
0:04:19 Like engineers are kind of taking the other side of like, no, it’s always going to be bad.
0:04:23 You’re just going to get worse and worse. Yeah. That’s called coping. I’m just going to put that
0:04:30 out there. That’s that’s called coping. Yeah. I mean, there’s very little that I’ve wanted to get
0:04:35 coded up that I haven’t been able to. I mean, sometimes it might take me like hours of going
0:04:41 back and forth and having to talk to perplexity to find answers or going to grok and ask questions to get
0:04:47 and you know, I’ll use other resources to try to help me overcome roadblocks, but there’s very
0:04:52 little that I haven’t been able to do that I’ve wanted to get coded up, which brings me to the topic
0:04:58 of today’s episode. I’ve actually been rebuilding the future tools website completely through vibe
0:05:03 coding. So I wanted to sort of break down my process a little bit, talk about the whole process
0:05:07 that I’ve been going through. We also want to talk a little bit about what Peter levels has been doing
0:05:13 with his airplane game because that’s been pretty fascinating to watch. Yeah. But let me start with like a little
0:05:22 bit of context. So I’ve built future tools back in 2022. So it’s been around for, you know, three ish, three
0:05:29 plus years now. And about a year and a half ago, I started the process of overhauling it and making it
0:05:36 better. And I started looking for developers. At first, I sort of cheaped out. This was back before I was
0:05:40 really making any money. I sort of cheaped out, went to Fiverr and found somebody that was really
0:05:45 good at Webflow and was trying to get them to sort of rebuild my site for me, you know, just by going
0:05:50 back and forth and trying to get them. But that guy flaked on me. He fixed a few things, but then
0:05:56 eventually just disappeared. I never heard from him again. Right. I probably over the whole, like maybe
0:06:01 three or four weeks I was working with him, paid him $800 just on like doing little tasks and trying to
0:06:05 fix things with me. Yeah. But that process was frustrating because he just sort of disappeared
0:06:10 on me at one point and then literally never heard from him again to this day. Hopefully he’s still
0:06:15 alive. I don’t know. Then I started the process of looking for a more, I guess, the legit developer,
0:06:20 not the people on Fiverr aren’t legit, but I wanted somebody with more of a track record. So I started
0:06:24 looking for recommendations. I found somebody that came highly recommended that had built some really
0:06:29 cool websites and software for other people, brought him on board to start working on it.
0:06:34 he mapped out this whole roadmap of what he was going to create. He actually made some new designs,
0:06:42 gave me a whole like branding sheet with color schemes and fonts and, you know, had a whole workup
0:06:49 for me. And I went, cool, that looks good. Let’s do it. Gave him a 50% deposit on the project. You know,
0:06:53 the final 50% would be given to him right around the time it was being completed before he handed over
0:06:59 all the assets. He started getting going with it. And then he sort of disappeared for like
0:07:04 months and months and months and was stopped responding to me. And every once in a while,
0:07:08 he would send me an update and be like, Hey, I’ve made a little bit of progress. Here’s what we’ve got.
0:07:14 And it was like barely any movement at all, like six months in. Yeah. And then eventually he asked for
0:07:19 final payment and was like, yeah, I’m really, really close. Can I get final payment? And then we’ll
0:07:24 finish this thing off. And I was like, all right, finally, we’re going to finally get this thing done.
0:07:30 I gave him the final payment. And then like, it was still barely done at all. I mean, he made some
0:07:36 progress, but not enough progress over the time. And then I started saying, Hey, here’s some of the
0:07:41 things that I wanted in the site. Like originally the site had a light mode and dark mode. Can you add
0:07:46 dark mode? And he was like, Oh, that’s out of scope. That’s going to cost you this much. And I was like,
0:07:51 but that was in the original plan. Like if you look at the original site, that was already there.
0:07:56 That should have been in the scope because that’s what we had. I don’t want to lose features that I
0:08:01 had. And he was like, okay, I’ll figure that out. I’ll just do it for you. And won’t charge you for
0:08:05 that. And I was like, okay, cool. And then I mentioned another feature that was in the original
0:08:08 site that he didn’t code into the new site. And he’s like, that’s out of scope.
0:08:13 This is typically how it is working with engineers. Like I’ve did several startups in Silicon Valley,
0:08:19 even with like more professional engineers, you run into the same thing. And I think this is why
0:08:22 vibe coding is going to get more and more popular too, is because did you already have these issues
0:08:25 before probably gonna be worse in the future? Cause they’re gonna be vibe code. You’re gonna be paying
0:08:31 them to do the vibe code. Might as well do it yourself. Yeah, exactly. And I mean, that’s what
0:08:36 ended up happening. I eventually just cut my losses and went, I’m not getting anywhere with this. I feel
0:08:41 like if I keep going with this person, it’s going to be years. So I cut them loose. I actually went
0:08:46 back to X and made a post. Hey, is there anybody out there that is really good at web flow? I’m just
0:08:49 going to keep it on web flow to keep things simple, but I need somebody to like redesign it and rebuild
0:08:54 it for me. Got a bunch of people to reach out and they all gave me like these insane bids. And I went, you
0:08:59 know what? I’m just going to go try to code this thing up myself. We talked to Riley Brown, go check that
0:09:03 episode out if you haven’t already. And we vibe coded some stuff with him. And I’m like, I bet I can
0:09:10 rebuild this myself. And I basically managed to clone all of the existing functionality of the future
0:09:17 tools website in like two days. And now I’m actually building new functionality on the site and actually
0:09:22 improving the design and stuff, which is why I haven’t pushed out the new site yet, but I managed to
0:09:28 clone what I’ve already got like super, super quick. And that’s sort of the context of why I decided to do
0:09:35 this. Like people are out there going, Oh, AI is taking jobs. But the thing is I really, really,
0:09:41 really struggled to actually pay people to do the job for me. Like it was really, really, really difficult
0:09:47 for me to give people money and have them do work that I was expecting. I was trying, I was trying to
0:09:53 not be the guy that goes, all right, I’m just going to do everything with AI. I was trying to redistribute
0:09:58 the wealth and be like, Hey, I’ll pay people to go do this for me. But it was a nightmare. So I
0:10:03 basically decided to go code it myself out of necessity because it was such a pain in the ass
0:10:08 working with other developers. Right. So anyway, that’s the whole context in a slight rant.
0:10:15 But what I wanted to do is sort of break down the process of how I did it, because I didn’t just like
0:10:19 go into clod and say, Hey, make this for me. I’ve actually been jumping around and using a lot of
0:10:22 different tools. So that’s kind of what I wanted to break down.
0:10:26 Have you been mainly using cursor? Like how have you been building this? How did you get started?
0:10:28 Like, how did you clone your existing website?
0:10:34 Yeah. So I’ve been doing cursor and then windsurf. I’ve been sort of switching back and forth to see
0:10:38 which one I like better, you know, cursor, I was running into problems. So I went to windsurf and
0:10:43 then I ran into problems with windsurf. So I jumped back to cursor and I’m paying for both of them right
0:10:48 now and jumping back and forth. But the initial starting point, here’s how I did it. I actually went to
0:10:55 Vercels V0.dev. And this site is actually really, really good at cloning existing sites and actually
0:11:01 improving the design for you. So I think I just fed it an image. I don’t even think I gave it a prompt.
0:11:06 I fed it an image. I literally took a screenshot of what future tools looks like, fed it into Vercel.
0:11:11 It looks like I didn’t even give it any text prompt at all with it. It was literally just a screenshot.
0:11:18 And then V0 went on to say, future tools like AI directory. I’ll recreate this AI tools directory
0:11:22 website with a clean modern design, similar to the one in the image. Let’s build a responsive
0:11:26 landing page with all the key components. And then you could see it went through and did its thing.
0:11:31 And what you’re seeing on the right was the design that it spit back out at me, which is a cleaner
0:11:37 looking design than what I have now on future tools. And I went, cool. That’s good bones for it.
0:11:40 That is so magical. We take all this for granted though, right? Like if you told somebody like
0:11:44 three years ago that like you could just do that, it would just make the website for you. It’d copy
0:11:49 your existing one, give you a new one, look better. And you just get the code for it. What are you
0:11:54 talking about? Yeah, this was first try, right? Like that’s crazy. This was me throwing in a screenshot
0:11:59 and then it went, all right, here’s a sort of nicer looking version of what you already have.
0:12:04 It’s just cleaner. It just looks more modern and clean, but similar structure, similar bones.
0:12:08 And all I did was screenshot it, throw it to Vercel and it gave me the code up here.
0:12:15 And then I’m able to download a zip file. So I downloaded the zip file here and it had to,
0:12:20 you know, these handful of files that it made for me. And I took the files that it made for me
0:12:25 and I threw them into cursor. You know, basically you create a folder and then you open that folder
0:12:29 inside of cursor. And so that was the sort of first step was, all right, I’ve got like the newer
0:12:35 cleaner design that I’d like. I used V zero for that. My next step, and this is something that you
0:12:41 and I have chatted a lot about is when you’re building with like AI coding tools, having like
0:12:50 context in a roadmap is a really sort of key way to do this. So what I did was I went to GPT 4.0,
0:12:54 but I literally gave it my existing URL. I said, Hey, I want to do a deep dive on everything.
0:13:00 The website futures tools.io does. I need a detailed breakdown of the entire site and all the current
0:13:05 existing functionality. So I literally gave it the URL and said, break down all the functionality that
0:13:10 I’ve already got on the site. Right. I want to know everything. It asked me a few questions.
0:13:16 I answered the questions and then you can see future tools.io website structure and functionality.
0:13:22 And it just wrote up this entire breakdown. So we’ve got a navigation and layout breakdown,
0:13:28 a user interaction breakdown, the search and filtering breakdown, but you can see it just broke down
0:13:35 everything for me in this like super, super detailed report here. So what I did was I pulled that whole
0:13:41 text document into a chat inside of cursor. And I literally asked cursor to give me a step-by-step
0:13:48 roadmap, like break this down into steps that we can complete. Yeah. I use clod 3.7 and then we can
0:13:54 see the roadmap that it created here. So future tools.io rebuild roadmap. And then here’s the full
0:13:58 on roadmap. It broke it down into nine phases. I think it’s funny how it says one week, one week,
0:14:03 three to five days, three to four weeks, two weeks. I’m already through like the first like six steps
0:14:07 and I did it in two days. Are you saving all that as like a text file or like a markdown file or
0:14:12 something like that? Or yeah, this is a file called roadmap.md. So you can see the roadmap here.
0:14:16 It’s really, really detailed. You can see, I still have a lot of steps that I want to complete on it,
0:14:20 but this is all new functionality. All of the existing functionality of future tools has already
0:14:26 been built. This is all brand new here, but now what I can do is whenever I give it a prompt,
0:14:32 I can come down here inside of cursor and there’s a little button that says add context. I can click on
0:14:38 that and I can just type roadmap and you can see it gives me the option to attach my roadmap.md file,
0:14:44 my markdown file with all of this entire roadmap here. So now every time I prompt, it has the context
0:14:50 of my roadmap with every prompt. So it knows where we are in the process and I’m actually checking off
0:14:55 stuff as it gets complete. So it knows what stuff has been done and what stuff it hasn’t gotten done.
0:15:00 And it’s using that context every single time I give it a prompt and it’s getting there.
0:15:04 Yeah. I think most people haven’t realized how powerful that is to have like a markdown file
0:15:08 or a text file with like your plans in it while you’re working with a code base. Cause then when
0:15:13 you use like the agent mode or whatever, the LLM automatically can look at that document to give
0:15:18 it better context about how to help you implement something. Yeah. Right. So here’s the site. I didn’t
0:15:23 load up every single tool in here right now. I only put 20 tools in here cause, um, the actual future
0:15:27 tool site has about 3,500 tools. And I’m like, that’s overkill for like the testing phase.
0:15:32 Right. Yeah. So I only put 20 tools in here to test, but as you can see, it’s got these tools here.
0:15:38 They’re tagged. They’ve got like a sort of liking system that you can put on them. It’s pulled in all
0:15:43 of the tool details, all of this filtering works. Now, are you still using the same backend or did you
0:15:47 have it create some kind of new backend and you ported the data over? No. So my old backend was all
0:15:52 webflow. Yeah. Right. So I built the whole system on webflow before. Yeah. This new one is all built
0:15:57 built on react. Yeah. And then it’s using super base for the database. Yeah. And then I’m using
0:16:03 wasabi for the CDN to host all of the images and, you know, uh, media files. Cool. So this one’s built
0:16:08 from the ground up. It’s not using webflow. It’s not using any of that kind of stuff. It’s all, you
0:16:14 know, react type script, super base and wasabi. My first goal was clone what I have. Right. So that’s
0:16:19 kind of what it did right now. There’s going to be a new section for gadgets, which is, you know,
0:16:23 the same as like the software tools, but it’s going to have like, you know, the rabbit R one and the
0:16:30 meta Ray bands and, you know, physical AI tools, but everything is functioning and working exactly like
0:16:36 it did on the other site right now. And I built this in two days. That’s insane. It’s got everything
0:16:41 in it. I’m just still sort of putting finishing touches on it and I built it in two days, but now
0:16:46 you’re in control of it. It’s actually yours. You can modify it however you want to. You also get the
0:16:50 extra benefit of every single time the LLMs get better. You’ll probably improve your site
0:16:54 automatically. Like a new version comes out and you can like, Hey, analyze my website. Look at my
0:16:58 roadmap. How could you improve it? I’ve noticed that every time they get better, like you just show it
0:17:03 the same code and it always discovers new things that can improve. Yeah, exactly. And you know,
0:17:09 the other thing is like, I was having my struggles with a developer and I posted on X like, Hey,
0:17:13 I’m rebuilding future tools. Do you guys have any ideas of like cool features to add? I’m sort of like
0:17:18 phishing for additional functionality that might make the site even cooler. And somebody said,
0:17:23 is it really that much easier than working with a developer? Don’t you run into like issues and bugs
0:17:30 with the vibe coding process? And my response to it was, yeah, you do run into issues, but you can
0:17:35 usually fix the issues within like an hour or two, where if you’re actually working with a developer
0:17:40 that doesn’t respond very quickly, it might be two weeks before you get like a small bug fixed.
0:17:45 Right. But now if I run into a bug and I’m like, Oh, that’s not good. I can typically fix it in like
0:17:50 10 minutes. Sometimes they’re a little more complex and it takes me like two hours and
0:17:55 Claude won’t do it inside of cursor. So I’ll go over to grok and ask grok, or I’ll go to like
0:18:00 perplexity and do a search on perplexity to see if I can find this solution. So sometimes I go to some
0:18:07 other tools to help me find the solution, but I usually get a solution within a couple hours at the
0:18:12 very, very longest versus going back and forth with a developer, trying to explain the problem
0:18:17 to them. Maybe they misunderstand what I’m trying to fix and they fix the wrong thing. And it’s just,
0:18:22 yeah, the struggles I have with vibe coding are way smaller than the struggles that I have with
0:18:27 actually hiring developers. Yeah. Like communicating with people, like what’s actually in your head is
0:18:32 very hard, like very, very difficult, almost impossible. I think we talked about this on another
0:18:36 episode about in game design, you know, Jonathan blows up about that, like the disconnect between
0:18:40 a designer and a business person and an engineer working on a game or any product. There’s always
0:18:44 a disconnect there. And it feels like with vibe coding, you can actually get closer to building
0:18:48 what was actually in your head. That was so hard to explain to other people. You almost have this
0:18:52 like connection with the AI where you’re going back and forth and you understand how it works and
0:18:56 understand what you’re wanting and that’s going to get better over time. Yeah. And all the issues you’re
0:19:00 talking about, like you’ve got bugs and things like that. It’s kind of annoying. I’ve noticed every time a new
0:19:04 model comes out, there’s less of that. Yeah. Like a year ago when I would try this and like you’d change
0:19:09 stuff, there’d be so many errors almost every single time that it was like, okay, I’m just not really
0:19:13 going to do this right now. It’s not really fun. But now it’s like the errors are much less like
0:19:18 anything I do. Like I went back to change my website, lore.com yesterday, just like add some stuff about
0:19:23 the next wave podcast and update some numbers and some other things like that. And I literally just like
0:19:29 talked to cursor for like two minutes and like then all of it was done. I’m like, oh, that’s crazy. I
0:19:32 thought that was going to be like a, you know, a few hour thing for me to go do all that. No,
0:19:36 it was literally two minutes and I was done. And the last time I tried that, it gave me errors. So
0:19:39 they’re getting better and better. And like the errors are gonna be less and less. And like,
0:19:42 you know, probably in a year or two, it’s like, you’re just gonna be talking to it and it’s gonna
0:19:47 be creating things and errors are gonna be rare. Right? Yeah. Another thing that other people have
0:19:53 said is like, well, you’re able to do this because you know about the platforms, you know, to ask it to
0:19:59 use react, you know, about Supabase for the database, you know, about places like Amazon S3 or Wasabi for
0:20:05 like hosting the media files. You know, you have that knowledge, so you know how to prompt it to get
0:20:12 that stuff. And that’s BS, right? Because what I did was I went to chat GPT and said, give me a roadmap.
0:20:18 Like, I don’t know anything about this stuff. Tell me what to build it with. Chat GPT is what
0:20:24 recommended using react. It actually recommended using Firebase for the database portion. But when
0:20:28 I was trying to set up Firebase, I was kind of running into some issues where the API wasn’t
0:20:33 connecting very well. It’s more expensive too. And it’s more expensive. And so I literally went to
0:20:39 perplexity and went, I’m having issues with Firebase. What are some alternatives? And it went, try Supabase.
0:20:45 And I went, okay, I’ll try that. So like, I did not know to tell it to use react. I did not know to
0:20:50 tell it to use Supabase. I did not know that stuff. Yeah, I think you’re making a great point,
0:20:54 though. This is like what in Silicon Valley, they started calling like people who are high agency,
0:20:59 people who will actually go do things versus just like, oh, I can’t do it. I think people have gotten
0:21:02 used to being like that, like having lower agency because things have been so hard. And if it’s
0:21:07 outside of their wheelhouse or their area of expertise, they’re like, well, I can’t do that. So
0:21:11 but with AI, that’s all changing. So like, people really should be like restructuring their
0:21:16 thinking to be like, higher agency of like, oh, I can do things because like the AI, whatever I
0:21:20 don’t know, it just fills that gap in now. Yeah, like, Matt, you didn’t know how to do database
0:21:23 stuff. I was gonna ask you, I was like, I bet you didn’t know anything about Supabase or how to set
0:21:27 it up. I’ve done that before. But like, I’m not an expert at it. But you know, it’s easier than
0:21:31 setting up your own like MySQL or whatever, but still, it’s kind of a pain. But now AI will just like
0:21:36 either do it for you or walk you through any of this stuff. Yeah, exactly. There’s this sort of
0:21:42 newer concept that came out that was released by Anthropic called MCP or model context protocol.
0:21:48 And before we hit record on this episode, you were saying, oh, it’ll be super nice when the AI can
0:21:54 just like look at your website for you and go, oh, this isn’t working very well. Let me fix that for
0:21:59 you. Or I noticed there’s some broken images on your site. Let me go and fix that for you. MCP fixes that
0:22:05 for you. So a lot of people go, well, isn’t MCP just like an API, right? So basically an API is
0:22:11 there’s software tools all over the internet that you can tie into with your own software, right? So
0:22:17 let’s say you want to use chat GPT and build chat GPT into your own product. Well, you can go and use
0:22:23 open AI’s API and make your tool talk to open AI’s tools. Or let’s say you want to integrate with
0:22:29 Google Calendar. You can actually make your tool connect to Google Calendar through an API, right?
0:22:37 It’s API is how programmers make their tool talk to other tools, right? Well, the problem with APIs
0:22:43 is every API is different. Every API has its own documentation. They’re formatted in different ways.
0:22:49 You call upon the APIs in different ways. Everybody sort of builds their APIs in different ways. It’s not
0:22:55 super, super standardized. The problem with that is when you ask a large language model to go and
0:23:02 integrate with an API, it basically has to go learn how to use that API for every tool it’s trying to
0:23:07 integrate with. You need to either feed it documentation or through trial and error, it will figure out how
0:23:12 to connect to that API. So let’s say you’ve got like a GitHub API. You’re trying to connect cursor with
0:23:20 GitHub and you have a Slack API and you have a local file system API, right? They’re all different
0:23:26 and cursor’s trying to communicate with each one and it needs to do it differently every single time.
0:23:32 Well, what MCP did is it added a different layer. It added another layer to sort of standardize it.
0:23:39 So now all GitHub has to do is create an MCP layer which communicates with its API. All Slack has to do
0:23:48 is create an MCP that communicates with its API. And now cursor only has to understand how MCPs work. It doesn’t
0:23:57 have to understand how every single API works. These companies can basically tie their API to the MCP and now
0:24:04 all of these LLMs can communicate easier because there’s a sort of standardization of the API connection.
0:24:07 I see. So when people are saying they’re like setting up an MCP server, they’re basically setting
0:24:12 up a server where they connect various APIs to it and show it how to use the APIs. And then the LLM just
0:24:15 pings the one server, like one API call basically.
0:24:21 Sort of. So basically, each one of these will have its own sort of MCP connection.
0:24:21 Yeah.
0:24:28 And basically, cursor can understand MCPs. It understands how to communicate with the MCP.
0:24:35 And GitHub would have designed their MCP to connect to their API. Slack would have designed their MCP to
0:24:41 connect to the API. So it’s a standardized way where the large language models only need to know how to
0:24:48 communicate with MCPs in general. And then each company will create an MCP to communicate with
0:24:49 their API, if that makes sense.
0:24:51 Yeah. I think I get it now.
0:24:59 Yeah. So in cursor, if I go up here to my settings, you can see that there’s an MCP setting here and you
0:25:04 can see that I’ve got an MCP for Superbase. I’ve got an MCP for browser tools and I’ve got an MCP
0:25:11 for GitHub. This basically means that when I’m communicating over here in my chat window, in my
0:25:20 agent window inside of cursor, I can tell it to do something like break down the schema of my current
0:25:25 database. Right. And I can give it that prompt. I’m going to switch it to chat mode so it doesn’t go and
0:25:32 try to write code on me. And what it will do is it will go and use this MCP to connect directly to my
0:25:40 Superbase account and look at the databases for me. So in the past, it would basically create SQL files
0:25:45 and tell you to copy this SQL file over into Superbase. And then if something wasn’t working,
0:25:49 it would look at the SQL file and try to figure it out. But it wasn’t actually looking at the database
0:25:54 for you. Well, now this is actually looking straight into the database for me and it can break
0:26:00 down the whole like schema for me. It can break down all of the various tables and all of the various
0:26:05 columns that are in each table and it can break it all down by literally looking directly in my database
0:26:13 for me. Same with the GitHub MCP. I gave it access to be able to look at my GitHub account. So instead of
0:26:18 having to write get commands, it can literally just push and commit stuff directly into GitHub.
0:26:24 without having to do like terminal commands and it can go and read my various files inside of GitHub.
0:26:31 So this is the GitHub for browser tools, MCP. It’s actually a Chrome extension. So because that Chrome
0:26:39 extension is turned on and I have the MCP installed over inside of cursor, what this will do is it gives
0:26:44 it the ability to look at all of your console logs, look at all of your network logs. So, you know,
0:26:50 like if you go to inspect and you have this like little sidebar here, you’ve got like console and
0:26:57 sources and, you know, memory and all of this kind of stuff that you can view. Well, this MCP gives
0:27:05 your LLM access to all of this. So it can go read console errors. It can go read network errors,
0:27:12 all of that kind of stuff, but it’s also capable of capturing screenshots. So I can give it a command.
0:27:17 I can go in there and I can say, make this change for me. And after you make the change,
0:27:23 take a screenshot of my webpage to make sure the change worked. And so it’ll go and do all the
0:27:28 coding, make all the changes, and then connect through the MCP, take a screenshot of my website.
0:27:33 And then if it looks right, it’ll say, good. It looks like the changes were implemented correctly.
0:27:38 If it doesn’t look right, it might say, oops, it looks like we’ve got some broken images and some
0:27:43 console errors. Let me fix that for you. And then it will just keep on processing. And then it’ll
0:27:47 process again and then write more code. And it’ll take another screenshot and look at the console
0:27:53 logs and go, oh, it looks like we fixed it now. It’s actually working. So with these model context
0:27:57 protocols, it can actually connect directly into my browser. Now look at the console logs,
0:28:02 take screenshots for me, look at my super base database for me, look at my GitHub account for
0:28:08 me and actually pull information from all of these places to sort of better inform the next decision
0:28:14 that it makes. And it just makes everything so much easier because it spots its own errors. Now
0:28:18 when it spots its own errors, it then goes, oops, not working. Let’s go ahead and fix it.
0:28:21 Right. Yeah. That seems like a game changer. Cause I was always the issue is like,
0:28:25 you’d create something with cursor, then it would make an error and it wouldn’t know. And now if I can
0:28:29 just look at the screen and look at the console log as well and find out that there was an error
0:28:30 and fix it.
0:28:35 Yeah. And more and more companies are creating MCPs to be able to connect with large language models.
0:28:41 So a Firebase has an MCP as well. Now, obviously Slack has one. We were just seeing that as one of
0:28:42 the examples. Yeah.
0:28:48 The GitHub MCP is really cool because I mean, you could have it go and look at a whole like
0:28:53 GitHub repository and have it break down what’s going on in that repository for you now.
0:28:58 Yeah. I did notice that like MCPs got really big on social media in the last week, but people were
0:29:02 saying like the concept’s been around for like a year now or something, right? Was it just like the
0:29:07 models were not good enough to actually be doing this or there was a not enough like MCPs out there that
0:29:08 were actually useful.
0:29:13 Here’s Anthropics website introducing the model context protocol. It came out November 25th,
0:29:21 2024. The big difference though was the IDEs didn’t have model concept protocol, like direct
0:29:22 integrations yet.
0:29:22 Yeah.
0:29:28 And you can actually write like some JavaScript code to make it connect, but you kind of needed
0:29:29 to know how to do that.
0:29:30 Right.
0:29:35 Now they’ve just implemented them into the IDEs like cursor and windsurf and tools like that.
0:29:41 They have like an MCP setting section now to make it a lot easier to connect. So there’s that
0:29:48 element. And the other element is that a lot of these tools had to create MCPs that connected to
0:29:54 their APIs, right? Like Superbase’s MCP just came out like two weeks ago. So we’ve only been able to
0:29:58 connect directly to Superbase using the MCP a couple of weeks ago.
0:30:01 So that makes sense. So it came out in November and then people are just now like
0:30:03 figuring out how to build all the software to make it work.
0:30:03 Yeah.
0:30:07 As soon as it came out, I checked and there’s one for a Unity actually. So like there’s like one for
0:30:08 like almost anything you can imagine.
0:30:14 Yeah. So this model context protocol for like the browser tools one that can take screenshots and do
0:30:17 that. We can see that it came out February 10th.
0:30:17 Okay.
0:30:23 It’s only been out for a month. So Superbase, you’ve only been able to connect via an MCP for two weeks
0:30:24 or so. Yeah.
0:30:28 The browser tools, you’ve only been able to connect to an MCP for a month now.
0:30:28 Right.
0:30:34 So that’s why it was never really a big deal before. It was like, it was introduced by Anthropic and
0:30:39 they’re like, Hey, this exists now. But all of the companies that had APIs that wanted to be able to
0:30:44 connect to large language models, then had to go and create these protocols so that people could call
0:30:50 upon them. And the IDEs needed to make it simpler to connect to the protocols. So that’s why it got big
0:30:53 recently is because they all just got way more accessible.
0:30:57 It seems like this is a big moment because like almost any kind of software you can use,
0:31:01 if they develop an MCP, you’ll be able to use it and interact with it through the LLMs now,
0:31:01 which is awesome.
0:31:05 Yeah. That’s why everybody’s going, Holy crap, this is pretty big.
0:31:05 Yeah.
0:31:11 Because like you said, you can connect it straight into Unity and it can probably push code into
0:31:12 Unity for you, you know?
0:31:17 I mean, so actually, you know, the LLMs already can push code into Unity, but the problem they have is
0:31:21 like setting up the scenes and different things like that in the game is something you do in the
0:31:27 editor. But apparently with the MCP now, you can actually create the scenes and different objects
0:31:32 in the editor through the MCP. Super excited to try that out.
0:31:32 Yeah.
0:31:37 Like we’ve been talking about this episode and last week’s episode, this stuff’s just getting easier
0:31:42 and easier. So it’s crazy how easy it’s gotten to be able to do this stuff. And the more the models
0:31:48 get smarter, like you mentioned, the less errors it creates in the code and the more tools we can
0:31:54 directly connect to the LLM, the better it’s able to tell if it’s doing a good job or not,
0:31:55 essentially. Right.
0:32:00 Yeah. And then combine all that with the models getting smarter and smarter. And then I guess
0:32:02 you got AGI or at least what I would probably call AGI.
0:32:08 Yeah. Yeah. I mean, it’s, it’s really wild, but yeah. So that’s how I’ve been building future tools.
0:32:13 That’s been the process. I think the really sort of valuable part that people should know is like,
0:32:18 you don’t really have to start from scratch, right? If, if you have a concept for a website
0:32:23 and there’s a design out there that you like, go and find a site that you like the design of that
0:32:30 sort of maybe has the functionality you want, pull that into V zero, let it clone it and use that as
0:32:34 your jumping off point. Like don’t make, you know, blatant copies of other people’s sites,
0:32:40 but use it as like a nice beginning template to start from and then use something like, you know,
0:32:46 Google’s deep research, or I even think GPT 4.5 will work well to write up a roadmap for you. I think it
0:32:51 would actually do a pretty good job, but you can use one of these tools to explain what you want to
0:32:57 create. Let it write up a roadmap for you. Let it tell you what tools to use, like react and super base
0:33:04 and all that kind of stuff. And you don’t have to figure this stuff out. The AI already knows the
0:33:08 best tools to accomplish what you’re trying to accomplish. So if you have your starting design
0:33:13 of what you want and you have a roadmap that, Hey, I helped you generate with the tech stack already
0:33:18 figured out for you. Like the rest is just putting it into cursor and telling cursor to follow the
0:33:23 roadmap. Right. Yeah. And especially now that there’s MCPs, I mean, it’s like, you’re gonna be able to
0:33:27 do beyond just coding, right? It’s gonna be vibe everything, right? It’ll vibe anything you want.
0:33:31 Cause it’s not gonna be limited to just coding anymore. Like I said, even like game design or
0:33:34 almost any kind of business, you’re gonna probably have the same thing where AI will be able to help
0:33:38 you do anything. Yeah. And it’s gonna be more and more important to be high agency. You know,
0:33:43 you don’t have an excuse anymore to not say, Oh, I don’t know how to do that. The AI will help you
0:33:47 figure out how to do it. Matt didn’t know how to do database stuff. He’s now a database expert
0:33:52 through vibe coding. Yeah. It’s wild. The other topic that I think would be cool to talk
0:33:58 about is Peter levels, right? He just built this like airplane game and it’s honestly not that
0:34:03 impressive looking. What’s impressive about it is that he’s just like one dude that vibe coded this
0:34:10 whole thing by himself. And he just passed a million dollars in monthly annual revenue from this one
0:34:15 person coding up this app by themselves. If you’re listening to this, instead of watching on YouTube,
0:34:20 it’s a very basic game he’s built. It’s like in a three GS, which is a framework that helps you
0:34:25 build 3d stuff in the browser. It’s a simple, you know, flight simulator where you fly around. But
0:34:28 the interesting thing about this, I think a lot of people are gonna look at like, Oh, this is just a
0:34:33 stupid thing. It’s a fad. The game itself, I would not doubt if it dies off like in a month or two.
0:34:39 But what’s very interesting is that this is like the ultimate example of vibe coding.
0:34:44 Yeah, it’s like a concept. He put like a really basic demo of it. And he just shared it online.
0:34:48 And he started talking with this community and like tweaking it based on stuff they were saying,
0:34:52 which when you were talking about building your future tools website, it kind of reminded me of that.
0:34:56 It’s like, Matt, you kind of revived building future tools with your community, right?
0:35:00 Yeah, where they were giving you feedback, and you’re building your roadmap based on what they said.
0:35:03 And then you just went off and built it. He did the same kind of thing with this game where he’s just
0:35:09 chatting with people online, his fans, about what should he add next. And now that it’s so easy that you can
0:35:12 actually just do that yourself versus going off and hiring an entire team, it takes all this time.
0:35:17 No, you just chat with the AI and you just change stuff and you just push it live and then you just
0:35:21 share it again and get more and more attention like that. And it’s almost like how people used to do
0:35:25 stuff on Twitch, right? Live streaming. I’m kind of convinced this is gonna be like a huge trend over
0:35:29 the next year or two, where you’re gonna have all these influencers realizing like, oh, wait a minute,
0:35:33 I could use my audience and I could just like live build something and it doesn’t be a game. It could be
0:35:38 software too. I’m gonna live build something and my audience is going to interact with me as I build it.
0:35:44 I’m going to make it for them. Yeah, it’s crazy, right? So I just pulled up this tweet from Peter
0:35:52 here. It says what you see 200 K MRR. So he’s at $200,000 a month in MRR. So that’s what 1.4 million
0:35:58 a year. If you’re extrapolating it out, what you don’t see, it’s only Stripe fees. That’s like the only
0:36:04 cost he has. Apparently no Apple fees. Cause it’s all web-based. He owns a hundred percent of it.
0:36:10 He’s spent $0 on ads. Yeah. Be your own influencer, no co-founders and happy customers.
0:36:15 Right. Right. Like that’s wild. And game designers are so pissed off. They’re all like tweeting about
0:36:20 it. Yeah. And I mean, it’s not a great looking game. I mean, it looks like a game that came out
0:36:26 in the nineties, the early nineties, but like the way he’s making money is see all the ads. They’re
0:36:32 like everywhere. He literally put an ad on planets and he put an ad on a blimp and he’s got ads on like
0:36:40 the buildings on the screen and he charges five grand for one of these ads. Yeah. A fake plane
0:36:45 like flies past really quick. I’m like, and he tweets out their names too. That’s another element
0:36:50 of it. Okay. Okay. He’s been tweeting out all of them. That’s probably the bigger benefit. Cause I
0:36:55 mean, the fact that they’re like ad is in this game, it’s not even direct response. Like people aren’t
0:37:01 like clicking on these ads to get to the website. It’s just pure branding. Right. I actually looked at it
0:37:04 cause I’m like, Oh, maybe I’ll throw an ad up there for future tools just for fun. Cause
0:37:09 who’s using it right now? Vibe coders and people interested in AI future tools would be great.
0:37:13 Yeah. And then I saw five grand a month to put my name on a blimp that planes are just flying by.
0:37:19 And I was like, eh, I don’t think I can justify that. But right. Yeah. And the people probably
0:37:23 stopped playing it in like a month or two, but still it just conceptually the fact that you can build
0:37:26 something and then interact with your community. Cause now it’s so easy to actually build something
0:37:30 in near real time like that. That’s a game changer that most people are not really fully
0:37:33 thinking through. I don’t think it’s gonna be limited to just games. You know, we’ve talked
0:37:37 about in the past, like, okay, there’s going to be software that you just build for yourself. It’s
0:37:41 like, yeah, but how many people are actually high agency and it will actually do that. But there may
0:37:46 be, you know, custom tailored software for your audience that you’re building for them. They get to
0:37:51 actually kind of see it as like almost entertainment that you’re building it for them. Like, Oh, cool.
0:37:56 Yeah. I want that feature. And then build it for them in the next like hour. Right. And push it out.
0:37:59 And then people are paying you. I think you’re going to see so many things like this where people
0:38:04 are going to make tons of money building games, software, other kinds of entertainment tools.
0:38:09 I know. This is wild. Yeah. So you actually get into the game and you have missions,
0:38:13 missions, shoot down all the balloons, shoot down other planes, fly to space.
0:38:18 It did not have any of this. Like when he started, he literally every feature has been people in the
0:38:21 community, just saying stuff like, why doesn’t it have that? He’s like, cool. Yeah. And then he’ll
0:38:26 like tweet out a link saying like, it’s up now, like a little video clip. So wild. So wild. That’s
0:38:30 been the whole process. It’s literally him just sharing little short videos on X and then people
0:38:36 saying, cool. Why doesn’t it have this? Yeah. We need more UFOs. We need a castle. Cool. It has that
0:38:41 now. Yeah. Yeah. Now it has it. You know, next he’s going to be charging for planes. Cool. Here’s a
0:38:46 stripe link. Here’s a link where you can buy planes. Buy the planes. It should have advertisements. Cool.
0:38:51 Here’s ads. And the entire process has been that of like him vibe coding it and then just pushing it
0:38:56 out to the community and doing it interactively. Yeah. It’s wild. I mean, we’re entering this crazy
0:39:03 world where literally anybody can code anything they want. They can build anything they want. If you have
0:39:08 an idea, if you’ve got a vision, go and use AI. You can use something like V0 to help you with the
0:39:14 original design. Use something like ChatGPT or Claude or Grok3 to go and help you map out the
0:39:19 product. The game part itself. I don’t think that’s like the main thing. I really think it’s idea of
0:39:23 like if you’re in a business right now and you have a new idea for a product and you have a big community
0:39:28 instead of making it super polished and like waiting years to build out the thing. Why doesn’t the one
0:39:33 business leader who has the idea literally just push it out to social media and share a prototype
0:39:37 and see if people have any interest and they say, Hey, why doesn’t it do this? Cool. Now it does
0:39:42 that. And like have an iterative process with your community to build software. I think that’s a huge
0:39:47 opportunity that people are not realizing. Yeah. I mean, essentially you’re right. Like that’s kind
0:39:50 of what I’m trying to do with the future tools website, right? I’m just going, Oh, right now I’m
0:39:56 building the bones, right? Like I need to get the sort of site up and live first. And then once it’s live
0:40:01 and people are actually accessing it, I’m going to do that. Hey, you should add this feature. Hey, you should
0:40:06 on the tools pages at this functionality. Hey, you should let me chat with natural language
0:40:10 and then just share it when it’s up. And then like, what should I do next? And get feedback and feed that
0:40:15 into the whole app and put it in your plan. And a hundred percent. That is my plan with it is to like
0:40:21 literally sort of let the community steer where the website goes. I just got to get that initial launch
0:40:26 done, right? I’m still finishing the finishing touches once it’s launched. That’s the goal. And I think
0:40:31 it’s also great marketing because if I go and say, what do you want to see on the site? And then somebody says,
0:40:36 Oh, you should add a natural language where I can just tell it my problem and it will recommend
0:40:42 tools. And then I go, cool. That was added. Yeah. They’re going to be stoked that I like actually
0:40:47 took their idea, built it in, and they’re going to help market my product for me. Yeah. Yeah. Like
0:40:53 thousand true fans on steroids, right? Exactly. Like imagine if a game like Grand Theft Auto did that,
0:40:58 right? Like, Oh my God, they actually built my feature in that I suggested. This is amazing. I need to go
0:41:03 tell everybody I’m a Grand Theft Auto fan for life now, you know? Yeah. But anyway, I think that’s a
0:41:08 good place to wrap this one up. I think, you know, this whole vibe coding concept, I’ve been nerding
0:41:13 out about it. I’ve actually been producing less YouTube videos on my own YouTube channel lately
0:41:18 because I’ve been so obsessed with just coding stuff. Future tools is just one thing that I’ve
0:41:23 been building. Like I’ve been building other little apps for my own like personal helpful use.
0:41:27 We’ve talked about this on a past episode. I love finding little bottlenecks in my business
0:41:32 and going, can I make a little app like to build the workflow to get past that bottleneck?
0:41:37 And I’ve made a whole bunch of little tools like that now, like a tool that easily converts any image
0:41:42 file to a single JPEG file. Doesn’t matter what type of file I drop in converts it instantly.
0:41:48 I created a tool with Whisper where I can take any of my short form videos, drop them in and it
0:41:53 instantly transcribes it. And then I can take that transcription and use it as my description over on
0:42:00 Instagram. So all of my Instagram sort of descriptions on everything is all written by AI now. I showed you
0:42:05 on a past video when we had Matthew Berman on, I showed you this tool that I’m calling the video
0:42:10 producer where I can throw a whole bunch of videos in and it helps me with ideas on new videos based on the
0:42:15 videos that are in my database. And so I’m just like finding bottlenecks or things that I want to
0:42:20 work through in my business and then using AI to build tools that solve those problems for me.
0:42:25 And I’m doing it as a one man show just by myself. Like I’m just, yeah, I have this idea. Let’s build
0:42:31 it. I’m not making as many videos, but like I’m obsessing over this right now. Yeah. I mean,
0:42:34 that’s great. A new problem of like, what do you focus on? You’re going to be more and more
0:42:38 infinite possibilities, right? With AI that it’ll be able to go off and do things for you. And it’s like,
0:42:41 what do I focus on? I became obsessed with my game. I’m like, well, the game may not even make
0:42:46 money. I’m like, you know, with AI and everything else, like what is it eventually even going to
0:42:49 matter? Like as long as I enjoy it, you know, I enjoy the process of making it.
0:42:54 I see a lot of this is like upfront work now to save me time in the future. Right. So like,
0:43:00 maybe, yeah, I’m producing one less video this week because I made a tool that improves my workflow.
0:43:05 But next week I have one less task on my plate because I improved this workflow. So it’s like,
0:43:11 I’m building all these little things that ideally will free me up in the future.
0:43:15 The problem is I keep on getting more ideas for things that will free me up in the future.
0:43:20 So I never actually end up freeing any time because I end up building the new ideas that I have. It’s a
0:43:21 weird place I’m in.
0:43:25 Have you ever considered just like open sourcing all that stuff that you’re building and just like,
0:43:29 yeah, yeah, yeah. I’ve open sourced a lot of them. Okay. Yeah. A lot of the little tools that
0:43:32 I’ve made, like the, any image to JPEG converter. I put that up on GitHub.
0:43:37 Have you told people on like X and YouTube or no? Yeah, I’ve mentioned it. I haven’t made a big
0:43:42 deal about it, but I have like mentioned it in the past. Like I’ll talk about a tool that I made and
0:43:46 people will be like, oh, that’s really cool. Can I get access to it? And I’ll be like, yep, it’s on
0:43:51 GitHub now. Go grab it. You know? So like I’ve pointed it out, but I haven’t really like went out
0:43:56 and megaphone announced it. Right. But yeah, as I’m building it, I’m putting everything on GitHub.
0:44:02 Once I feel like it’s ready for other people to go use, then I make it public on GitHub and I set the
0:44:06 license to an open source license. So anybody can go and use it. I don’t think I’ll open source future
0:44:11 tools. There’s already too many clones of my website. I don’t know if I want to like enable more of that,
0:44:14 but yeah, pretty much everything else I build, I open source.
0:44:19 So what’s the key takeaway for people? Be high agency and you don’t have excuses now and go build
0:44:24 stuff. All of that. Yeah. No excuses. You know, you get an idea, you have bottlenecks in your business,
0:44:28 go build workflows to fix them. You want to go build a new website. You don’t need that developer anymore.
0:44:35 You need a new business model. Maybe you can monetize your fans by collaboratively building with
0:44:41 them and working with their ideas and helping convert them into your marketing force because you’re
0:44:45 building with the community. Yeah. On that note, I think we should go ahead and wrap this one up.
0:44:52 We’re definitely going to be talking a lot more about AI and coding and vibe coding and all of this
0:44:57 cool stuff in future episodes. So if you’re not already, make sure you’re subscribed either on YouTube
0:45:02 or wherever you listen to podcasts, preferably on both. Watch us on video and then enjoy it so much
0:45:06 that you need to go listen to it again in your car. Just subscribe everywhere. We’d really appreciate
0:45:11 that. And we’ll make sure more awesome episodes like this keep showing up for you. Thank you so
0:45:15 much for tuning in and we’ll see you in the next episode. Yeah. Thank you.
0:00:12 today we’re talking vibe coding. In this episode, I’m going to break down exactly how I’ve been
0:00:17 rebuilding the Future Tools website completely with AI and how you can build your own website.
0:00:21 We’re going to break down a whole new business model that’s emerging because of these AI tools
0:00:28 and show you a business that’s doing $1.4 million a year from one single creator. I’m going to
0:00:34 break down what MCPs are, what APIs are, what all of the acronyms are, and we’re going to deep dive
0:00:42 into how you can build your own businesses and products and different tools using this AI vibe
0:00:47 coding concept. So let’s go ahead and jump right in. Today, we’re going to talk about vibe coding,
0:00:55 one of the hottest buzzwords in the AI world, and it’s all about vibing and coding, I guess.
0:00:58 How would you define vibe coding, Nathan?
0:01:03 Actually, I know Andre Karpathy. He’s the one who coined it. I think of actually Riley Brown,
0:01:07 because he was the first person I saw talking about vibe coding, really. He was sharing all
0:01:11 this stuff online. We’re like, hey, I’m just chatting with Whisperflow, and then it’s transcribing
0:01:17 what I say into text, and I feed that into cursor, and then it makes something. And he’s kind of
0:01:21 like describing that whole process, how different that is versus you sit down and you got to code
0:01:25 everything by hand. You got to really think it through. It’s a different process. So when I think
0:01:29 of vibe coding, I think, you know, in theory, it’s a more relaxed process, right? You could even,
0:01:34 in theory, be doing something else. You know, like in the evenings now, I find that when my wife is like
0:01:38 taking her shower and getting other stuff ready and other things, I’ll be like hanging out on the couch,
0:01:41 maybe watching something on YouTube, educational and vibe coding.
0:01:46 Dude, I do the same thing. Yeah. So this is like become a common thing for me. We’re like now in
0:01:50 time periods where I probably wouldn’t have been productive in the past. I’m actually productive
0:01:54 during those time periods. Like I’m like watching some educational and making something. And it’s
0:01:58 actually kind of fun to make stuff now because you can just chat with your computer and it does the
0:02:03 thing. Yeah. I mean, I usually do it in my office. I got a sort of double monitor set up here,
0:02:08 right? So typically what I’m doing is on my main monitor right in front of me. I’ve got,
0:02:12 you know, my coding app open my IDE. I’m usually either using cursor or windsurf. I sort of switch
0:02:17 back and forth. We’ll talk about that in a few minutes here, but I’m got my coding going here
0:02:23 and I’ve got YouTube typically opened up on my right screen and I’ll give it a prompt of what I
0:02:27 want it to go code for me and then hit enter. And then I’ll press play on my YouTube video,
0:02:32 sort of watch for a minute. Notice out of the corner of my eye that the coding has sort of finished
0:02:36 pause my YouTube video. All right, let’s check out what you did. Right. All right. It’s not quite
0:02:42 working right. Go fix this. Enter press play on my YouTube video again, go back to watching it.
0:02:47 And so it’s almost like you can code passively, right? Or, you know, I’ll do the same thing.
0:02:51 I might hang out in my living room, grab my laptop and just be sitting there on my laptop doing some
0:02:57 coding while the TV’s going. But yeah, I mean, essentially I think vibe coding is using natural
0:03:02 language to write code. I actually looked it up on perplexity to see how perplexity would define it
0:03:08 based on its research. And it says coined by Andre Carpathy in February, 2025. This method involves
0:03:13 describing software ideas in natural language, allowing AI to write the code and iteratively
0:03:20 refining it through conversation. So basically just talking to a coding app saying, here’s what I want
0:03:25 you to create. Go do it. Letting it go and write the code. You double check it and either saying, cool,
0:03:30 that worked or no, that didn’t work, go fix it. And just going back and forth using natural language
0:03:35 until eventually you get the app that you wanted to get out of it.
0:03:40 Right. And there’s been a whole conversation on X around this too. Like engineers are really skeptical
0:03:44 of vibe coding and like turning it into a meme, like in a negative way. They’re talking about like vibe
0:03:48 coding, you know, the systems for like how planes are flown. Like probably don’t want that, you know?
0:03:54 Yeah, I agree. Right. I think there’s definitely critical like infrastructure areas where you don’t
0:03:59 want to hear that it was vibe coded. Right. Yeah. I don’t want the software that the air traffic
0:04:03 controllers use to be vibe coded. Right. Right. Right. And a lot of times right now, the AI, when it does
0:04:07 make things, there are typically like small bugs and there’s probably things that are not the way that
0:04:11 you would have hand coded if you were like an expert. I kind of believe over time, those things are going to
0:04:15 just get fixed. Like the models are going to get better. You know, it’s definitely a big debate on X right now.
0:04:19 Like engineers are kind of taking the other side of like, no, it’s always going to be bad.
0:04:23 You’re just going to get worse and worse. Yeah. That’s called coping. I’m just going to put that
0:04:30 out there. That’s that’s called coping. Yeah. I mean, there’s very little that I’ve wanted to get
0:04:35 coded up that I haven’t been able to. I mean, sometimes it might take me like hours of going
0:04:41 back and forth and having to talk to perplexity to find answers or going to grok and ask questions to get
0:04:47 and you know, I’ll use other resources to try to help me overcome roadblocks, but there’s very
0:04:52 little that I haven’t been able to do that I’ve wanted to get coded up, which brings me to the topic
0:04:58 of today’s episode. I’ve actually been rebuilding the future tools website completely through vibe
0:05:03 coding. So I wanted to sort of break down my process a little bit, talk about the whole process
0:05:07 that I’ve been going through. We also want to talk a little bit about what Peter levels has been doing
0:05:13 with his airplane game because that’s been pretty fascinating to watch. Yeah. But let me start with like a little
0:05:22 bit of context. So I’ve built future tools back in 2022. So it’s been around for, you know, three ish, three
0:05:29 plus years now. And about a year and a half ago, I started the process of overhauling it and making it
0:05:36 better. And I started looking for developers. At first, I sort of cheaped out. This was back before I was
0:05:40 really making any money. I sort of cheaped out, went to Fiverr and found somebody that was really
0:05:45 good at Webflow and was trying to get them to sort of rebuild my site for me, you know, just by going
0:05:50 back and forth and trying to get them. But that guy flaked on me. He fixed a few things, but then
0:05:56 eventually just disappeared. I never heard from him again. Right. I probably over the whole, like maybe
0:06:01 three or four weeks I was working with him, paid him $800 just on like doing little tasks and trying to
0:06:05 fix things with me. Yeah. But that process was frustrating because he just sort of disappeared
0:06:10 on me at one point and then literally never heard from him again to this day. Hopefully he’s still
0:06:15 alive. I don’t know. Then I started the process of looking for a more, I guess, the legit developer,
0:06:20 not the people on Fiverr aren’t legit, but I wanted somebody with more of a track record. So I started
0:06:24 looking for recommendations. I found somebody that came highly recommended that had built some really
0:06:29 cool websites and software for other people, brought him on board to start working on it.
0:06:34 he mapped out this whole roadmap of what he was going to create. He actually made some new designs,
0:06:42 gave me a whole like branding sheet with color schemes and fonts and, you know, had a whole workup
0:06:49 for me. And I went, cool, that looks good. Let’s do it. Gave him a 50% deposit on the project. You know,
0:06:53 the final 50% would be given to him right around the time it was being completed before he handed over
0:06:59 all the assets. He started getting going with it. And then he sort of disappeared for like
0:07:04 months and months and months and was stopped responding to me. And every once in a while,
0:07:08 he would send me an update and be like, Hey, I’ve made a little bit of progress. Here’s what we’ve got.
0:07:14 And it was like barely any movement at all, like six months in. Yeah. And then eventually he asked for
0:07:19 final payment and was like, yeah, I’m really, really close. Can I get final payment? And then we’ll
0:07:24 finish this thing off. And I was like, all right, finally, we’re going to finally get this thing done.
0:07:30 I gave him the final payment. And then like, it was still barely done at all. I mean, he made some
0:07:36 progress, but not enough progress over the time. And then I started saying, Hey, here’s some of the
0:07:41 things that I wanted in the site. Like originally the site had a light mode and dark mode. Can you add
0:07:46 dark mode? And he was like, Oh, that’s out of scope. That’s going to cost you this much. And I was like,
0:07:51 but that was in the original plan. Like if you look at the original site, that was already there.
0:07:56 That should have been in the scope because that’s what we had. I don’t want to lose features that I
0:08:01 had. And he was like, okay, I’ll figure that out. I’ll just do it for you. And won’t charge you for
0:08:05 that. And I was like, okay, cool. And then I mentioned another feature that was in the original
0:08:08 site that he didn’t code into the new site. And he’s like, that’s out of scope.
0:08:13 This is typically how it is working with engineers. Like I’ve did several startups in Silicon Valley,
0:08:19 even with like more professional engineers, you run into the same thing. And I think this is why
0:08:22 vibe coding is going to get more and more popular too, is because did you already have these issues
0:08:25 before probably gonna be worse in the future? Cause they’re gonna be vibe code. You’re gonna be paying
0:08:31 them to do the vibe code. Might as well do it yourself. Yeah, exactly. And I mean, that’s what
0:08:36 ended up happening. I eventually just cut my losses and went, I’m not getting anywhere with this. I feel
0:08:41 like if I keep going with this person, it’s going to be years. So I cut them loose. I actually went
0:08:46 back to X and made a post. Hey, is there anybody out there that is really good at web flow? I’m just
0:08:49 going to keep it on web flow to keep things simple, but I need somebody to like redesign it and rebuild
0:08:54 it for me. Got a bunch of people to reach out and they all gave me like these insane bids. And I went, you
0:08:59 know what? I’m just going to go try to code this thing up myself. We talked to Riley Brown, go check that
0:09:03 episode out if you haven’t already. And we vibe coded some stuff with him. And I’m like, I bet I can
0:09:10 rebuild this myself. And I basically managed to clone all of the existing functionality of the future
0:09:17 tools website in like two days. And now I’m actually building new functionality on the site and actually
0:09:22 improving the design and stuff, which is why I haven’t pushed out the new site yet, but I managed to
0:09:28 clone what I’ve already got like super, super quick. And that’s sort of the context of why I decided to do
0:09:35 this. Like people are out there going, Oh, AI is taking jobs. But the thing is I really, really,
0:09:41 really struggled to actually pay people to do the job for me. Like it was really, really, really difficult
0:09:47 for me to give people money and have them do work that I was expecting. I was trying, I was trying to
0:09:53 not be the guy that goes, all right, I’m just going to do everything with AI. I was trying to redistribute
0:09:58 the wealth and be like, Hey, I’ll pay people to go do this for me. But it was a nightmare. So I
0:10:03 basically decided to go code it myself out of necessity because it was such a pain in the ass
0:10:08 working with other developers. Right. So anyway, that’s the whole context in a slight rant.
0:10:15 But what I wanted to do is sort of break down the process of how I did it, because I didn’t just like
0:10:19 go into clod and say, Hey, make this for me. I’ve actually been jumping around and using a lot of
0:10:22 different tools. So that’s kind of what I wanted to break down.
0:10:26 Have you been mainly using cursor? Like how have you been building this? How did you get started?
0:10:28 Like, how did you clone your existing website?
0:10:34 Yeah. So I’ve been doing cursor and then windsurf. I’ve been sort of switching back and forth to see
0:10:38 which one I like better, you know, cursor, I was running into problems. So I went to windsurf and
0:10:43 then I ran into problems with windsurf. So I jumped back to cursor and I’m paying for both of them right
0:10:48 now and jumping back and forth. But the initial starting point, here’s how I did it. I actually went to
0:10:55 Vercels V0.dev. And this site is actually really, really good at cloning existing sites and actually
0:11:01 improving the design for you. So I think I just fed it an image. I don’t even think I gave it a prompt.
0:11:06 I fed it an image. I literally took a screenshot of what future tools looks like, fed it into Vercel.
0:11:11 It looks like I didn’t even give it any text prompt at all with it. It was literally just a screenshot.
0:11:18 And then V0 went on to say, future tools like AI directory. I’ll recreate this AI tools directory
0:11:22 website with a clean modern design, similar to the one in the image. Let’s build a responsive
0:11:26 landing page with all the key components. And then you could see it went through and did its thing.
0:11:31 And what you’re seeing on the right was the design that it spit back out at me, which is a cleaner
0:11:37 looking design than what I have now on future tools. And I went, cool. That’s good bones for it.
0:11:40 That is so magical. We take all this for granted though, right? Like if you told somebody like
0:11:44 three years ago that like you could just do that, it would just make the website for you. It’d copy
0:11:49 your existing one, give you a new one, look better. And you just get the code for it. What are you
0:11:54 talking about? Yeah, this was first try, right? Like that’s crazy. This was me throwing in a screenshot
0:11:59 and then it went, all right, here’s a sort of nicer looking version of what you already have.
0:12:04 It’s just cleaner. It just looks more modern and clean, but similar structure, similar bones.
0:12:08 And all I did was screenshot it, throw it to Vercel and it gave me the code up here.
0:12:15 And then I’m able to download a zip file. So I downloaded the zip file here and it had to,
0:12:20 you know, these handful of files that it made for me. And I took the files that it made for me
0:12:25 and I threw them into cursor. You know, basically you create a folder and then you open that folder
0:12:29 inside of cursor. And so that was the sort of first step was, all right, I’ve got like the newer
0:12:35 cleaner design that I’d like. I used V zero for that. My next step, and this is something that you
0:12:41 and I have chatted a lot about is when you’re building with like AI coding tools, having like
0:12:50 context in a roadmap is a really sort of key way to do this. So what I did was I went to GPT 4.0,
0:12:54 but I literally gave it my existing URL. I said, Hey, I want to do a deep dive on everything.
0:13:00 The website futures tools.io does. I need a detailed breakdown of the entire site and all the current
0:13:05 existing functionality. So I literally gave it the URL and said, break down all the functionality that
0:13:10 I’ve already got on the site. Right. I want to know everything. It asked me a few questions.
0:13:16 I answered the questions and then you can see future tools.io website structure and functionality.
0:13:22 And it just wrote up this entire breakdown. So we’ve got a navigation and layout breakdown,
0:13:28 a user interaction breakdown, the search and filtering breakdown, but you can see it just broke down
0:13:35 everything for me in this like super, super detailed report here. So what I did was I pulled that whole
0:13:41 text document into a chat inside of cursor. And I literally asked cursor to give me a step-by-step
0:13:48 roadmap, like break this down into steps that we can complete. Yeah. I use clod 3.7 and then we can
0:13:54 see the roadmap that it created here. So future tools.io rebuild roadmap. And then here’s the full
0:13:58 on roadmap. It broke it down into nine phases. I think it’s funny how it says one week, one week,
0:14:03 three to five days, three to four weeks, two weeks. I’m already through like the first like six steps
0:14:07 and I did it in two days. Are you saving all that as like a text file or like a markdown file or
0:14:12 something like that? Or yeah, this is a file called roadmap.md. So you can see the roadmap here.
0:14:16 It’s really, really detailed. You can see, I still have a lot of steps that I want to complete on it,
0:14:20 but this is all new functionality. All of the existing functionality of future tools has already
0:14:26 been built. This is all brand new here, but now what I can do is whenever I give it a prompt,
0:14:32 I can come down here inside of cursor and there’s a little button that says add context. I can click on
0:14:38 that and I can just type roadmap and you can see it gives me the option to attach my roadmap.md file,
0:14:44 my markdown file with all of this entire roadmap here. So now every time I prompt, it has the context
0:14:50 of my roadmap with every prompt. So it knows where we are in the process and I’m actually checking off
0:14:55 stuff as it gets complete. So it knows what stuff has been done and what stuff it hasn’t gotten done.
0:15:00 And it’s using that context every single time I give it a prompt and it’s getting there.
0:15:04 Yeah. I think most people haven’t realized how powerful that is to have like a markdown file
0:15:08 or a text file with like your plans in it while you’re working with a code base. Cause then when
0:15:13 you use like the agent mode or whatever, the LLM automatically can look at that document to give
0:15:18 it better context about how to help you implement something. Yeah. Right. So here’s the site. I didn’t
0:15:23 load up every single tool in here right now. I only put 20 tools in here cause, um, the actual future
0:15:27 tool site has about 3,500 tools. And I’m like, that’s overkill for like the testing phase.
0:15:32 Right. Yeah. So I only put 20 tools in here to test, but as you can see, it’s got these tools here.
0:15:38 They’re tagged. They’ve got like a sort of liking system that you can put on them. It’s pulled in all
0:15:43 of the tool details, all of this filtering works. Now, are you still using the same backend or did you
0:15:47 have it create some kind of new backend and you ported the data over? No. So my old backend was all
0:15:52 webflow. Yeah. Right. So I built the whole system on webflow before. Yeah. This new one is all built
0:15:57 built on react. Yeah. And then it’s using super base for the database. Yeah. And then I’m using
0:16:03 wasabi for the CDN to host all of the images and, you know, uh, media files. Cool. So this one’s built
0:16:08 from the ground up. It’s not using webflow. It’s not using any of that kind of stuff. It’s all, you
0:16:14 know, react type script, super base and wasabi. My first goal was clone what I have. Right. So that’s
0:16:19 kind of what it did right now. There’s going to be a new section for gadgets, which is, you know,
0:16:23 the same as like the software tools, but it’s going to have like, you know, the rabbit R one and the
0:16:30 meta Ray bands and, you know, physical AI tools, but everything is functioning and working exactly like
0:16:36 it did on the other site right now. And I built this in two days. That’s insane. It’s got everything
0:16:41 in it. I’m just still sort of putting finishing touches on it and I built it in two days, but now
0:16:46 you’re in control of it. It’s actually yours. You can modify it however you want to. You also get the
0:16:50 extra benefit of every single time the LLMs get better. You’ll probably improve your site
0:16:54 automatically. Like a new version comes out and you can like, Hey, analyze my website. Look at my
0:16:58 roadmap. How could you improve it? I’ve noticed that every time they get better, like you just show it
0:17:03 the same code and it always discovers new things that can improve. Yeah, exactly. And you know,
0:17:09 the other thing is like, I was having my struggles with a developer and I posted on X like, Hey,
0:17:13 I’m rebuilding future tools. Do you guys have any ideas of like cool features to add? I’m sort of like
0:17:18 phishing for additional functionality that might make the site even cooler. And somebody said,
0:17:23 is it really that much easier than working with a developer? Don’t you run into like issues and bugs
0:17:30 with the vibe coding process? And my response to it was, yeah, you do run into issues, but you can
0:17:35 usually fix the issues within like an hour or two, where if you’re actually working with a developer
0:17:40 that doesn’t respond very quickly, it might be two weeks before you get like a small bug fixed.
0:17:45 Right. But now if I run into a bug and I’m like, Oh, that’s not good. I can typically fix it in like
0:17:50 10 minutes. Sometimes they’re a little more complex and it takes me like two hours and
0:17:55 Claude won’t do it inside of cursor. So I’ll go over to grok and ask grok, or I’ll go to like
0:18:00 perplexity and do a search on perplexity to see if I can find this solution. So sometimes I go to some
0:18:07 other tools to help me find the solution, but I usually get a solution within a couple hours at the
0:18:12 very, very longest versus going back and forth with a developer, trying to explain the problem
0:18:17 to them. Maybe they misunderstand what I’m trying to fix and they fix the wrong thing. And it’s just,
0:18:22 yeah, the struggles I have with vibe coding are way smaller than the struggles that I have with
0:18:27 actually hiring developers. Yeah. Like communicating with people, like what’s actually in your head is
0:18:32 very hard, like very, very difficult, almost impossible. I think we talked about this on another
0:18:36 episode about in game design, you know, Jonathan blows up about that, like the disconnect between
0:18:40 a designer and a business person and an engineer working on a game or any product. There’s always
0:18:44 a disconnect there. And it feels like with vibe coding, you can actually get closer to building
0:18:48 what was actually in your head. That was so hard to explain to other people. You almost have this
0:18:52 like connection with the AI where you’re going back and forth and you understand how it works and
0:18:56 understand what you’re wanting and that’s going to get better over time. Yeah. And all the issues you’re
0:19:00 talking about, like you’ve got bugs and things like that. It’s kind of annoying. I’ve noticed every time a new
0:19:04 model comes out, there’s less of that. Yeah. Like a year ago when I would try this and like you’d change
0:19:09 stuff, there’d be so many errors almost every single time that it was like, okay, I’m just not really
0:19:13 going to do this right now. It’s not really fun. But now it’s like the errors are much less like
0:19:18 anything I do. Like I went back to change my website, lore.com yesterday, just like add some stuff about
0:19:23 the next wave podcast and update some numbers and some other things like that. And I literally just like
0:19:29 talked to cursor for like two minutes and like then all of it was done. I’m like, oh, that’s crazy. I
0:19:32 thought that was going to be like a, you know, a few hour thing for me to go do all that. No,
0:19:36 it was literally two minutes and I was done. And the last time I tried that, it gave me errors. So
0:19:39 they’re getting better and better. And like the errors are gonna be less and less. And like,
0:19:42 you know, probably in a year or two, it’s like, you’re just gonna be talking to it and it’s gonna
0:19:47 be creating things and errors are gonna be rare. Right? Yeah. Another thing that other people have
0:19:53 said is like, well, you’re able to do this because you know about the platforms, you know, to ask it to
0:19:59 use react, you know, about Supabase for the database, you know, about places like Amazon S3 or Wasabi for
0:20:05 like hosting the media files. You know, you have that knowledge, so you know how to prompt it to get
0:20:12 that stuff. And that’s BS, right? Because what I did was I went to chat GPT and said, give me a roadmap.
0:20:18 Like, I don’t know anything about this stuff. Tell me what to build it with. Chat GPT is what
0:20:24 recommended using react. It actually recommended using Firebase for the database portion. But when
0:20:28 I was trying to set up Firebase, I was kind of running into some issues where the API wasn’t
0:20:33 connecting very well. It’s more expensive too. And it’s more expensive. And so I literally went to
0:20:39 perplexity and went, I’m having issues with Firebase. What are some alternatives? And it went, try Supabase.
0:20:45 And I went, okay, I’ll try that. So like, I did not know to tell it to use react. I did not know to
0:20:50 tell it to use Supabase. I did not know that stuff. Yeah, I think you’re making a great point,
0:20:54 though. This is like what in Silicon Valley, they started calling like people who are high agency,
0:20:59 people who will actually go do things versus just like, oh, I can’t do it. I think people have gotten
0:21:02 used to being like that, like having lower agency because things have been so hard. And if it’s
0:21:07 outside of their wheelhouse or their area of expertise, they’re like, well, I can’t do that. So
0:21:11 but with AI, that’s all changing. So like, people really should be like restructuring their
0:21:16 thinking to be like, higher agency of like, oh, I can do things because like the AI, whatever I
0:21:20 don’t know, it just fills that gap in now. Yeah, like, Matt, you didn’t know how to do database
0:21:23 stuff. I was gonna ask you, I was like, I bet you didn’t know anything about Supabase or how to set
0:21:27 it up. I’ve done that before. But like, I’m not an expert at it. But you know, it’s easier than
0:21:31 setting up your own like MySQL or whatever, but still, it’s kind of a pain. But now AI will just like
0:21:36 either do it for you or walk you through any of this stuff. Yeah, exactly. There’s this sort of
0:21:42 newer concept that came out that was released by Anthropic called MCP or model context protocol.
0:21:48 And before we hit record on this episode, you were saying, oh, it’ll be super nice when the AI can
0:21:54 just like look at your website for you and go, oh, this isn’t working very well. Let me fix that for
0:21:59 you. Or I noticed there’s some broken images on your site. Let me go and fix that for you. MCP fixes that
0:22:05 for you. So a lot of people go, well, isn’t MCP just like an API, right? So basically an API is
0:22:11 there’s software tools all over the internet that you can tie into with your own software, right? So
0:22:17 let’s say you want to use chat GPT and build chat GPT into your own product. Well, you can go and use
0:22:23 open AI’s API and make your tool talk to open AI’s tools. Or let’s say you want to integrate with
0:22:29 Google Calendar. You can actually make your tool connect to Google Calendar through an API, right?
0:22:37 It’s API is how programmers make their tool talk to other tools, right? Well, the problem with APIs
0:22:43 is every API is different. Every API has its own documentation. They’re formatted in different ways.
0:22:49 You call upon the APIs in different ways. Everybody sort of builds their APIs in different ways. It’s not
0:22:55 super, super standardized. The problem with that is when you ask a large language model to go and
0:23:02 integrate with an API, it basically has to go learn how to use that API for every tool it’s trying to
0:23:07 integrate with. You need to either feed it documentation or through trial and error, it will figure out how
0:23:12 to connect to that API. So let’s say you’ve got like a GitHub API. You’re trying to connect cursor with
0:23:20 GitHub and you have a Slack API and you have a local file system API, right? They’re all different
0:23:26 and cursor’s trying to communicate with each one and it needs to do it differently every single time.
0:23:32 Well, what MCP did is it added a different layer. It added another layer to sort of standardize it.
0:23:39 So now all GitHub has to do is create an MCP layer which communicates with its API. All Slack has to do
0:23:48 is create an MCP that communicates with its API. And now cursor only has to understand how MCPs work. It doesn’t
0:23:57 have to understand how every single API works. These companies can basically tie their API to the MCP and now
0:24:04 all of these LLMs can communicate easier because there’s a sort of standardization of the API connection.
0:24:07 I see. So when people are saying they’re like setting up an MCP server, they’re basically setting
0:24:12 up a server where they connect various APIs to it and show it how to use the APIs. And then the LLM just
0:24:15 pings the one server, like one API call basically.
0:24:21 Sort of. So basically, each one of these will have its own sort of MCP connection.
0:24:21 Yeah.
0:24:28 And basically, cursor can understand MCPs. It understands how to communicate with the MCP.
0:24:35 And GitHub would have designed their MCP to connect to their API. Slack would have designed their MCP to
0:24:41 connect to the API. So it’s a standardized way where the large language models only need to know how to
0:24:48 communicate with MCPs in general. And then each company will create an MCP to communicate with
0:24:49 their API, if that makes sense.
0:24:51 Yeah. I think I get it now.
0:24:59 Yeah. So in cursor, if I go up here to my settings, you can see that there’s an MCP setting here and you
0:25:04 can see that I’ve got an MCP for Superbase. I’ve got an MCP for browser tools and I’ve got an MCP
0:25:11 for GitHub. This basically means that when I’m communicating over here in my chat window, in my
0:25:20 agent window inside of cursor, I can tell it to do something like break down the schema of my current
0:25:25 database. Right. And I can give it that prompt. I’m going to switch it to chat mode so it doesn’t go and
0:25:32 try to write code on me. And what it will do is it will go and use this MCP to connect directly to my
0:25:40 Superbase account and look at the databases for me. So in the past, it would basically create SQL files
0:25:45 and tell you to copy this SQL file over into Superbase. And then if something wasn’t working,
0:25:49 it would look at the SQL file and try to figure it out. But it wasn’t actually looking at the database
0:25:54 for you. Well, now this is actually looking straight into the database for me and it can break
0:26:00 down the whole like schema for me. It can break down all of the various tables and all of the various
0:26:05 columns that are in each table and it can break it all down by literally looking directly in my database
0:26:13 for me. Same with the GitHub MCP. I gave it access to be able to look at my GitHub account. So instead of
0:26:18 having to write get commands, it can literally just push and commit stuff directly into GitHub.
0:26:24 without having to do like terminal commands and it can go and read my various files inside of GitHub.
0:26:31 So this is the GitHub for browser tools, MCP. It’s actually a Chrome extension. So because that Chrome
0:26:39 extension is turned on and I have the MCP installed over inside of cursor, what this will do is it gives
0:26:44 it the ability to look at all of your console logs, look at all of your network logs. So, you know,
0:26:50 like if you go to inspect and you have this like little sidebar here, you’ve got like console and
0:26:57 sources and, you know, memory and all of this kind of stuff that you can view. Well, this MCP gives
0:27:05 your LLM access to all of this. So it can go read console errors. It can go read network errors,
0:27:12 all of that kind of stuff, but it’s also capable of capturing screenshots. So I can give it a command.
0:27:17 I can go in there and I can say, make this change for me. And after you make the change,
0:27:23 take a screenshot of my webpage to make sure the change worked. And so it’ll go and do all the
0:27:28 coding, make all the changes, and then connect through the MCP, take a screenshot of my website.
0:27:33 And then if it looks right, it’ll say, good. It looks like the changes were implemented correctly.
0:27:38 If it doesn’t look right, it might say, oops, it looks like we’ve got some broken images and some
0:27:43 console errors. Let me fix that for you. And then it will just keep on processing. And then it’ll
0:27:47 process again and then write more code. And it’ll take another screenshot and look at the console
0:27:53 logs and go, oh, it looks like we fixed it now. It’s actually working. So with these model context
0:27:57 protocols, it can actually connect directly into my browser. Now look at the console logs,
0:28:02 take screenshots for me, look at my super base database for me, look at my GitHub account for
0:28:08 me and actually pull information from all of these places to sort of better inform the next decision
0:28:14 that it makes. And it just makes everything so much easier because it spots its own errors. Now
0:28:18 when it spots its own errors, it then goes, oops, not working. Let’s go ahead and fix it.
0:28:21 Right. Yeah. That seems like a game changer. Cause I was always the issue is like,
0:28:25 you’d create something with cursor, then it would make an error and it wouldn’t know. And now if I can
0:28:29 just look at the screen and look at the console log as well and find out that there was an error
0:28:30 and fix it.
0:28:35 Yeah. And more and more companies are creating MCPs to be able to connect with large language models.
0:28:41 So a Firebase has an MCP as well. Now, obviously Slack has one. We were just seeing that as one of
0:28:42 the examples. Yeah.
0:28:48 The GitHub MCP is really cool because I mean, you could have it go and look at a whole like
0:28:53 GitHub repository and have it break down what’s going on in that repository for you now.
0:28:58 Yeah. I did notice that like MCPs got really big on social media in the last week, but people were
0:29:02 saying like the concept’s been around for like a year now or something, right? Was it just like the
0:29:07 models were not good enough to actually be doing this or there was a not enough like MCPs out there that
0:29:08 were actually useful.
0:29:13 Here’s Anthropics website introducing the model context protocol. It came out November 25th,
0:29:21 2024. The big difference though was the IDEs didn’t have model concept protocol, like direct
0:29:22 integrations yet.
0:29:22 Yeah.
0:29:28 And you can actually write like some JavaScript code to make it connect, but you kind of needed
0:29:29 to know how to do that.
0:29:30 Right.
0:29:35 Now they’ve just implemented them into the IDEs like cursor and windsurf and tools like that.
0:29:41 They have like an MCP setting section now to make it a lot easier to connect. So there’s that
0:29:48 element. And the other element is that a lot of these tools had to create MCPs that connected to
0:29:54 their APIs, right? Like Superbase’s MCP just came out like two weeks ago. So we’ve only been able to
0:29:58 connect directly to Superbase using the MCP a couple of weeks ago.
0:30:01 So that makes sense. So it came out in November and then people are just now like
0:30:03 figuring out how to build all the software to make it work.
0:30:03 Yeah.
0:30:07 As soon as it came out, I checked and there’s one for a Unity actually. So like there’s like one for
0:30:08 like almost anything you can imagine.
0:30:14 Yeah. So this model context protocol for like the browser tools one that can take screenshots and do
0:30:17 that. We can see that it came out February 10th.
0:30:17 Okay.
0:30:23 It’s only been out for a month. So Superbase, you’ve only been able to connect via an MCP for two weeks
0:30:24 or so. Yeah.
0:30:28 The browser tools, you’ve only been able to connect to an MCP for a month now.
0:30:28 Right.
0:30:34 So that’s why it was never really a big deal before. It was like, it was introduced by Anthropic and
0:30:39 they’re like, Hey, this exists now. But all of the companies that had APIs that wanted to be able to
0:30:44 connect to large language models, then had to go and create these protocols so that people could call
0:30:50 upon them. And the IDEs needed to make it simpler to connect to the protocols. So that’s why it got big
0:30:53 recently is because they all just got way more accessible.
0:30:57 It seems like this is a big moment because like almost any kind of software you can use,
0:31:01 if they develop an MCP, you’ll be able to use it and interact with it through the LLMs now,
0:31:01 which is awesome.
0:31:05 Yeah. That’s why everybody’s going, Holy crap, this is pretty big.
0:31:05 Yeah.
0:31:11 Because like you said, you can connect it straight into Unity and it can probably push code into
0:31:12 Unity for you, you know?
0:31:17 I mean, so actually, you know, the LLMs already can push code into Unity, but the problem they have is
0:31:21 like setting up the scenes and different things like that in the game is something you do in the
0:31:27 editor. But apparently with the MCP now, you can actually create the scenes and different objects
0:31:32 in the editor through the MCP. Super excited to try that out.
0:31:32 Yeah.
0:31:37 Like we’ve been talking about this episode and last week’s episode, this stuff’s just getting easier
0:31:42 and easier. So it’s crazy how easy it’s gotten to be able to do this stuff. And the more the models
0:31:48 get smarter, like you mentioned, the less errors it creates in the code and the more tools we can
0:31:54 directly connect to the LLM, the better it’s able to tell if it’s doing a good job or not,
0:31:55 essentially. Right.
0:32:00 Yeah. And then combine all that with the models getting smarter and smarter. And then I guess
0:32:02 you got AGI or at least what I would probably call AGI.
0:32:08 Yeah. Yeah. I mean, it’s, it’s really wild, but yeah. So that’s how I’ve been building future tools.
0:32:13 That’s been the process. I think the really sort of valuable part that people should know is like,
0:32:18 you don’t really have to start from scratch, right? If, if you have a concept for a website
0:32:23 and there’s a design out there that you like, go and find a site that you like the design of that
0:32:30 sort of maybe has the functionality you want, pull that into V zero, let it clone it and use that as
0:32:34 your jumping off point. Like don’t make, you know, blatant copies of other people’s sites,
0:32:40 but use it as like a nice beginning template to start from and then use something like, you know,
0:32:46 Google’s deep research, or I even think GPT 4.5 will work well to write up a roadmap for you. I think it
0:32:51 would actually do a pretty good job, but you can use one of these tools to explain what you want to
0:32:57 create. Let it write up a roadmap for you. Let it tell you what tools to use, like react and super base
0:33:04 and all that kind of stuff. And you don’t have to figure this stuff out. The AI already knows the
0:33:08 best tools to accomplish what you’re trying to accomplish. So if you have your starting design
0:33:13 of what you want and you have a roadmap that, Hey, I helped you generate with the tech stack already
0:33:18 figured out for you. Like the rest is just putting it into cursor and telling cursor to follow the
0:33:23 roadmap. Right. Yeah. And especially now that there’s MCPs, I mean, it’s like, you’re gonna be able to
0:33:27 do beyond just coding, right? It’s gonna be vibe everything, right? It’ll vibe anything you want.
0:33:31 Cause it’s not gonna be limited to just coding anymore. Like I said, even like game design or
0:33:34 almost any kind of business, you’re gonna probably have the same thing where AI will be able to help
0:33:38 you do anything. Yeah. And it’s gonna be more and more important to be high agency. You know,
0:33:43 you don’t have an excuse anymore to not say, Oh, I don’t know how to do that. The AI will help you
0:33:47 figure out how to do it. Matt didn’t know how to do database stuff. He’s now a database expert
0:33:52 through vibe coding. Yeah. It’s wild. The other topic that I think would be cool to talk
0:33:58 about is Peter levels, right? He just built this like airplane game and it’s honestly not that
0:34:03 impressive looking. What’s impressive about it is that he’s just like one dude that vibe coded this
0:34:10 whole thing by himself. And he just passed a million dollars in monthly annual revenue from this one
0:34:15 person coding up this app by themselves. If you’re listening to this, instead of watching on YouTube,
0:34:20 it’s a very basic game he’s built. It’s like in a three GS, which is a framework that helps you
0:34:25 build 3d stuff in the browser. It’s a simple, you know, flight simulator where you fly around. But
0:34:28 the interesting thing about this, I think a lot of people are gonna look at like, Oh, this is just a
0:34:33 stupid thing. It’s a fad. The game itself, I would not doubt if it dies off like in a month or two.
0:34:39 But what’s very interesting is that this is like the ultimate example of vibe coding.
0:34:44 Yeah, it’s like a concept. He put like a really basic demo of it. And he just shared it online.
0:34:48 And he started talking with this community and like tweaking it based on stuff they were saying,
0:34:52 which when you were talking about building your future tools website, it kind of reminded me of that.
0:34:56 It’s like, Matt, you kind of revived building future tools with your community, right?
0:35:00 Yeah, where they were giving you feedback, and you’re building your roadmap based on what they said.
0:35:03 And then you just went off and built it. He did the same kind of thing with this game where he’s just
0:35:09 chatting with people online, his fans, about what should he add next. And now that it’s so easy that you can
0:35:12 actually just do that yourself versus going off and hiring an entire team, it takes all this time.
0:35:17 No, you just chat with the AI and you just change stuff and you just push it live and then you just
0:35:21 share it again and get more and more attention like that. And it’s almost like how people used to do
0:35:25 stuff on Twitch, right? Live streaming. I’m kind of convinced this is gonna be like a huge trend over
0:35:29 the next year or two, where you’re gonna have all these influencers realizing like, oh, wait a minute,
0:35:33 I could use my audience and I could just like live build something and it doesn’t be a game. It could be
0:35:38 software too. I’m gonna live build something and my audience is going to interact with me as I build it.
0:35:44 I’m going to make it for them. Yeah, it’s crazy, right? So I just pulled up this tweet from Peter
0:35:52 here. It says what you see 200 K MRR. So he’s at $200,000 a month in MRR. So that’s what 1.4 million
0:35:58 a year. If you’re extrapolating it out, what you don’t see, it’s only Stripe fees. That’s like the only
0:36:04 cost he has. Apparently no Apple fees. Cause it’s all web-based. He owns a hundred percent of it.
0:36:10 He’s spent $0 on ads. Yeah. Be your own influencer, no co-founders and happy customers.
0:36:15 Right. Right. Like that’s wild. And game designers are so pissed off. They’re all like tweeting about
0:36:20 it. Yeah. And I mean, it’s not a great looking game. I mean, it looks like a game that came out
0:36:26 in the nineties, the early nineties, but like the way he’s making money is see all the ads. They’re
0:36:32 like everywhere. He literally put an ad on planets and he put an ad on a blimp and he’s got ads on like
0:36:40 the buildings on the screen and he charges five grand for one of these ads. Yeah. A fake plane
0:36:45 like flies past really quick. I’m like, and he tweets out their names too. That’s another element
0:36:50 of it. Okay. Okay. He’s been tweeting out all of them. That’s probably the bigger benefit. Cause I
0:36:55 mean, the fact that they’re like ad is in this game, it’s not even direct response. Like people aren’t
0:37:01 like clicking on these ads to get to the website. It’s just pure branding. Right. I actually looked at it
0:37:04 cause I’m like, Oh, maybe I’ll throw an ad up there for future tools just for fun. Cause
0:37:09 who’s using it right now? Vibe coders and people interested in AI future tools would be great.
0:37:13 Yeah. And then I saw five grand a month to put my name on a blimp that planes are just flying by.
0:37:19 And I was like, eh, I don’t think I can justify that. But right. Yeah. And the people probably
0:37:23 stopped playing it in like a month or two, but still it just conceptually the fact that you can build
0:37:26 something and then interact with your community. Cause now it’s so easy to actually build something
0:37:30 in near real time like that. That’s a game changer that most people are not really fully
0:37:33 thinking through. I don’t think it’s gonna be limited to just games. You know, we’ve talked
0:37:37 about in the past, like, okay, there’s going to be software that you just build for yourself. It’s
0:37:41 like, yeah, but how many people are actually high agency and it will actually do that. But there may
0:37:46 be, you know, custom tailored software for your audience that you’re building for them. They get to
0:37:51 actually kind of see it as like almost entertainment that you’re building it for them. Like, Oh, cool.
0:37:56 Yeah. I want that feature. And then build it for them in the next like hour. Right. And push it out.
0:37:59 And then people are paying you. I think you’re going to see so many things like this where people
0:38:04 are going to make tons of money building games, software, other kinds of entertainment tools.
0:38:09 I know. This is wild. Yeah. So you actually get into the game and you have missions,
0:38:13 missions, shoot down all the balloons, shoot down other planes, fly to space.
0:38:18 It did not have any of this. Like when he started, he literally every feature has been people in the
0:38:21 community, just saying stuff like, why doesn’t it have that? He’s like, cool. Yeah. And then he’ll
0:38:26 like tweet out a link saying like, it’s up now, like a little video clip. So wild. So wild. That’s
0:38:30 been the whole process. It’s literally him just sharing little short videos on X and then people
0:38:36 saying, cool. Why doesn’t it have this? Yeah. We need more UFOs. We need a castle. Cool. It has that
0:38:41 now. Yeah. Yeah. Now it has it. You know, next he’s going to be charging for planes. Cool. Here’s a
0:38:46 stripe link. Here’s a link where you can buy planes. Buy the planes. It should have advertisements. Cool.
0:38:51 Here’s ads. And the entire process has been that of like him vibe coding it and then just pushing it
0:38:56 out to the community and doing it interactively. Yeah. It’s wild. I mean, we’re entering this crazy
0:39:03 world where literally anybody can code anything they want. They can build anything they want. If you have
0:39:08 an idea, if you’ve got a vision, go and use AI. You can use something like V0 to help you with the
0:39:14 original design. Use something like ChatGPT or Claude or Grok3 to go and help you map out the
0:39:19 product. The game part itself. I don’t think that’s like the main thing. I really think it’s idea of
0:39:23 like if you’re in a business right now and you have a new idea for a product and you have a big community
0:39:28 instead of making it super polished and like waiting years to build out the thing. Why doesn’t the one
0:39:33 business leader who has the idea literally just push it out to social media and share a prototype
0:39:37 and see if people have any interest and they say, Hey, why doesn’t it do this? Cool. Now it does
0:39:42 that. And like have an iterative process with your community to build software. I think that’s a huge
0:39:47 opportunity that people are not realizing. Yeah. I mean, essentially you’re right. Like that’s kind
0:39:50 of what I’m trying to do with the future tools website, right? I’m just going, Oh, right now I’m
0:39:56 building the bones, right? Like I need to get the sort of site up and live first. And then once it’s live
0:40:01 and people are actually accessing it, I’m going to do that. Hey, you should add this feature. Hey, you should
0:40:06 on the tools pages at this functionality. Hey, you should let me chat with natural language
0:40:10 and then just share it when it’s up. And then like, what should I do next? And get feedback and feed that
0:40:15 into the whole app and put it in your plan. And a hundred percent. That is my plan with it is to like
0:40:21 literally sort of let the community steer where the website goes. I just got to get that initial launch
0:40:26 done, right? I’m still finishing the finishing touches once it’s launched. That’s the goal. And I think
0:40:31 it’s also great marketing because if I go and say, what do you want to see on the site? And then somebody says,
0:40:36 Oh, you should add a natural language where I can just tell it my problem and it will recommend
0:40:42 tools. And then I go, cool. That was added. Yeah. They’re going to be stoked that I like actually
0:40:47 took their idea, built it in, and they’re going to help market my product for me. Yeah. Yeah. Like
0:40:53 thousand true fans on steroids, right? Exactly. Like imagine if a game like Grand Theft Auto did that,
0:40:58 right? Like, Oh my God, they actually built my feature in that I suggested. This is amazing. I need to go
0:41:03 tell everybody I’m a Grand Theft Auto fan for life now, you know? Yeah. But anyway, I think that’s a
0:41:08 good place to wrap this one up. I think, you know, this whole vibe coding concept, I’ve been nerding
0:41:13 out about it. I’ve actually been producing less YouTube videos on my own YouTube channel lately
0:41:18 because I’ve been so obsessed with just coding stuff. Future tools is just one thing that I’ve
0:41:23 been building. Like I’ve been building other little apps for my own like personal helpful use.
0:41:27 We’ve talked about this on a past episode. I love finding little bottlenecks in my business
0:41:32 and going, can I make a little app like to build the workflow to get past that bottleneck?
0:41:37 And I’ve made a whole bunch of little tools like that now, like a tool that easily converts any image
0:41:42 file to a single JPEG file. Doesn’t matter what type of file I drop in converts it instantly.
0:41:48 I created a tool with Whisper where I can take any of my short form videos, drop them in and it
0:41:53 instantly transcribes it. And then I can take that transcription and use it as my description over on
0:42:00 Instagram. So all of my Instagram sort of descriptions on everything is all written by AI now. I showed you
0:42:05 on a past video when we had Matthew Berman on, I showed you this tool that I’m calling the video
0:42:10 producer where I can throw a whole bunch of videos in and it helps me with ideas on new videos based on the
0:42:15 videos that are in my database. And so I’m just like finding bottlenecks or things that I want to
0:42:20 work through in my business and then using AI to build tools that solve those problems for me.
0:42:25 And I’m doing it as a one man show just by myself. Like I’m just, yeah, I have this idea. Let’s build
0:42:31 it. I’m not making as many videos, but like I’m obsessing over this right now. Yeah. I mean,
0:42:34 that’s great. A new problem of like, what do you focus on? You’re going to be more and more
0:42:38 infinite possibilities, right? With AI that it’ll be able to go off and do things for you. And it’s like,
0:42:41 what do I focus on? I became obsessed with my game. I’m like, well, the game may not even make
0:42:46 money. I’m like, you know, with AI and everything else, like what is it eventually even going to
0:42:49 matter? Like as long as I enjoy it, you know, I enjoy the process of making it.
0:42:54 I see a lot of this is like upfront work now to save me time in the future. Right. So like,
0:43:00 maybe, yeah, I’m producing one less video this week because I made a tool that improves my workflow.
0:43:05 But next week I have one less task on my plate because I improved this workflow. So it’s like,
0:43:11 I’m building all these little things that ideally will free me up in the future.
0:43:15 The problem is I keep on getting more ideas for things that will free me up in the future.
0:43:20 So I never actually end up freeing any time because I end up building the new ideas that I have. It’s a
0:43:21 weird place I’m in.
0:43:25 Have you ever considered just like open sourcing all that stuff that you’re building and just like,
0:43:29 yeah, yeah, yeah. I’ve open sourced a lot of them. Okay. Yeah. A lot of the little tools that
0:43:32 I’ve made, like the, any image to JPEG converter. I put that up on GitHub.
0:43:37 Have you told people on like X and YouTube or no? Yeah, I’ve mentioned it. I haven’t made a big
0:43:42 deal about it, but I have like mentioned it in the past. Like I’ll talk about a tool that I made and
0:43:46 people will be like, oh, that’s really cool. Can I get access to it? And I’ll be like, yep, it’s on
0:43:51 GitHub now. Go grab it. You know? So like I’ve pointed it out, but I haven’t really like went out
0:43:56 and megaphone announced it. Right. But yeah, as I’m building it, I’m putting everything on GitHub.
0:44:02 Once I feel like it’s ready for other people to go use, then I make it public on GitHub and I set the
0:44:06 license to an open source license. So anybody can go and use it. I don’t think I’ll open source future
0:44:11 tools. There’s already too many clones of my website. I don’t know if I want to like enable more of that,
0:44:14 but yeah, pretty much everything else I build, I open source.
0:44:19 So what’s the key takeaway for people? Be high agency and you don’t have excuses now and go build
0:44:24 stuff. All of that. Yeah. No excuses. You know, you get an idea, you have bottlenecks in your business,
0:44:28 go build workflows to fix them. You want to go build a new website. You don’t need that developer anymore.
0:44:35 You need a new business model. Maybe you can monetize your fans by collaboratively building with
0:44:41 them and working with their ideas and helping convert them into your marketing force because you’re
0:44:45 building with the community. Yeah. On that note, I think we should go ahead and wrap this one up.
0:44:52 We’re definitely going to be talking a lot more about AI and coding and vibe coding and all of this
0:44:57 cool stuff in future episodes. So if you’re not already, make sure you’re subscribed either on YouTube
0:45:02 or wherever you listen to podcasts, preferably on both. Watch us on video and then enjoy it so much
0:45:06 that you need to go listen to it again in your car. Just subscribe everywhere. We’d really appreciate
0:45:11 that. And we’ll make sure more awesome episodes like this keep showing up for you. Thank you so
0:45:15 much for tuning in and we’ll see you in the next episode. Yeah. Thank you.
Episode 51: Is it really possible to rebuild an entire website using A.I.? Matt Wolfe (https://x.com/mreflow) and Nathan Lands (https://x.com/NathanLands) dive into the evolving world of AI-driven development, sharing their insights on the latest buzzword, vibe coding.
In this episode, Matt discusses how he is rebuilding the Future Tools website from scratch using AI, detailing the new business model emerging from these AI tools. They take listeners through the journey of leveraging tools like V0.dev, Cursor, and Windsurf to streamline the coding process, and explore how AI can help overcome challenges even with little to no prior coding experience. With AI taking center stage, the hosts delve into how it’s revolutionizing their workflows, the concept of MCP, and the real-world application of vibe coding, like the successful venture of Pieter Levels’ airplane game.
Check out The Next Wave YouTube Channel if you want to see Matt and Nathan on screen: https://lnk.to/thenextwavepd
—
Show Notes:
- (00:00) Vibe Coding & AI Business
- (04:50) Future Tools Rebuild and Updates
- (09:29) AI Over Human Workers: A Necessity
- (12:20) Deep Dive into Website Functions
- (14:08 Detailed Roadmap Integration Guide
- (17:01) Self-Coding vs. Developer Challenges
- (20:48) AI Empowering High Agency Mindset
- (24:52) Cursor MCPs: Enhanced Database Interaction
- (27:04) Automated Webpage Change Validation
- (32:03) Start with Existing Designs
- (34:56) Live Audience-Driven Development Trend
- (39:15) Iterative Community-Driven Product Development
- (41:39) Instant Video Transcription Tool
- (44:43) Subscribe for Future AI Episodes
—
Mentions:
- FutureTools: https://www.futuretools.io/
- V0:https://v0.dev/
- Cursor: https://www.cursor.com/
- Windsurf: https://codeium.com/windsurf
- Pieter Levels: https://x.com/levelsio
Get the guide to build your own Custom GPT: https://clickhubspot.com/tnw
—
Check Out Matt’s Stuff:
• Future Tools – https://futuretools.beehiiv.com/
• Blog – https://www.mattwolfe.com/
• YouTube- https://www.youtube.com/@mreflow
—
Check Out Nathan’s Stuff:
- Newsletter: https://news.lore.com/
- Blog – https://lore.com/
The Next Wave is a HubSpot Original Podcast // Brought to you by Hubspot Media // Production by Darren Clarke // Editing by Ezra Bakker Trupiano