Watch this 100x developer use Codex… it’s insane

I left on Tropic. I built a demo on Magipath. I think I got like a million views on that tweet and then I raised money in a week after that. >> So you prefer Codex over cloud code? I know. >> So I haven't touched cloud code in like 5 months. The only thing you can bet is the Mars are getting better. The future of work is going to be less about doing the thing but more about supervising the thing. This is Pro the founder of Magic Path. Vietro also used to work at Enthropic is very close to the top people at OpenAI and has access to AI models and agents that me and you don't have access to. In this podcast, we talk about why he stopped using clot code, how AI agents are going to eat all software, why he uses codeex to do everything and which AI company he would build today if he had to start from scratch. This interview is full of alpha. So, if you're serious about AI, you have to watch until the end. This is the David Andre podcast. Enjoy. Right, Petro? So, you mentioned that tools like Codex and Cloud Core are going to eat software. Can we start with you showing how you're currently building with Codex and then let's jump into more of the what future of the AI agent world looks like. >> Yes. So, here I have Codeex, right? And there's a couple tricks that I do when I work with Codex. I show you in a second, but first I want to show you like what I mean by Codex is going to eat software. So, as you know, Codex has a browser, right? That you can access. You can go to the side panel and then you can just have your files your terminal review and the browser. So what we build we basically build the functionality so that our product which is magic path can actually interact with these agents. What I meant by Codex is going to eat software. This is what I mean because now Codex is one of the biggest and strongest aress right so when the user now are using like their coding agent you know they already have their files they have their design MD they have their skills people now have their way of working what's missing is that sort of like place where the agents and humans can work together and so now you can see >> interface basically. >> Exactly. Yeah. >> So you prefer codex over cloud code right now? >> Yes. So I haven't touched cloud code in like five months. There's two reason why I think codeex is better. One the arness is better. I think they build a better agentic loop than cloud code. Cloud code is like too bloated. But I think the most important thing is like how much more token consumption the consumption of codeex I think is way better than cloud code. You can see like >> yeah I mean this on deepsw s like on the benchmarks you can see it you know codex is crushing cloud code and uh as you said for the same task it uses less tokens. Yeah, I mean this if this was hus once it will take forever and by the way right now I'm like on a high like so it's like you know if I use low of course it would have been much much faster but if this was hus it will take you to take forever it will cost a lot of money. >> What about front end though because like a lot of people prefer using Opus or even Gemini for front end because OpenAI you know it's great at large G factors but I think the front end is still weakness. >> Well you can see it's a it's a context problem. So look how much better this is made by codeex. Look how much better those designs are, right? >> And the reason is because when we run it through magic path, we provide some context to the model to act as a designer. >> I want to touch on more on like how it's eating software because not everybody sees in the future like you do, you know, and you have obviously like connections with people who are really on the cutting edge and and beyond. So like explain how agents like Codex or Cloud Code or Hermes or Pi are eating software like do you think like SAS business models will die? How people should think about starting a business? >> Yes, that's a great question. So I think there's two reason why uh software is dying and and codeex is eating everything. People now are moving to a direction where they really like to have their own process and they really like to have their own context, right? Think about this. In the past, let's say that you were working in a team, right? And you were so there was like a team, there was like an engineer, a PM, a designer, right? The amount of tooling that those people had access to was very first of all there was not a lot of tools and it was very limited to whatever activity or work this person was doing. You were an engineer, you were on VS Code. You were a PM, you just live in Google Drive. You were a designer, you were living in Figma. But now because of AI agents, not only the roles are collapsing. So a designer can be an engineer, engineer can be a PM, a PM can be an engineer. Everybody's now coming up with their own tooling, right? It's like, oh, some people like cloud code, some people like cursor, some people like codeex. And because of that, it's much harder now to say to somebody, this is the tool you have to use because they're so used to know how to run their own agent. And so for me, my only advice if if I was starting this company from scratch, roughly a year ago, it was literally like a year ago, maybe like two days ago. If I was starting this from scratch again, I would say well you know build this so it can be integrated with cursor can be integrated with codeex can be integrated with cloud code because the idea of going to a website to do something I think it's going to die and so to answer your original question how you can succeed in this environment the only thing that matters is brand and community. It's like you really need to have the community that R is behind you and say like this is what I want to the the platform that I want to use when it comes to interacting with those agents basically. >> By the way, here's the problem with every AI agent you build. It can't see the live web. It's reasoning over data that's months out of date which limits its power. Oxyab fixes that. Their web scraper API pulls structured real-time data from almost any public site and it ships with an official MCB server. So you can easily connect it to cloud code, cursor, codeex in a matter of minutes. Now your agent can scrape Amazon listings, Google results or real estate data just by you telling it in plain English. Oxyabs also have a headless browser built for AI agents. It handles JavaScript heavy apps and georestricted sites across 175 million residential IPs in more than 190 countries. That's why PCMAC named it the best proxy service of 2026. And what's nice is that Oxyabs gives you 2,000 scraped results completely for free. So go to oxyabs.io/david. Oh, and when you need to upgrade, use code David for 20% off on any plan. Thank you to Oxyabs for sponsoring this video. I see. So basically um you you're a big believer that like in the future most of the software will be used by agents and each person will have their own personal agent that they know has this necessary context about them use they use their favorite model their favorite harness whatever. >> Exactly. And the browser becomes the place where this interaction happens and but the browser stops being the main surface. It just becomes the secondary surface into which you communicate to those agents. Like for example, you know what I think is going to happen? I actually I can bet you billions of dollars of this. What I think is gonna happen is that OpenAI is gonna launch an SDK, right? It's gonna be like let's call it the codeex SDK where you're going to have basically like you know let's say like right now for example magic path for how powerful it is is still limited by the fact that is leaning in the browser. If you use our own agent and you select something, you can see the agent knows that this is selected, right? The problem is that when I do that, this doesn't show up as a pill into the the browser here, right? And so ideally what you want is that potentially codeex becomes this SDK where I can maybe take like some events on the browser and I can say like, oh, uh, send these events from the browser to the chat. And I really believe that's what's going to happen where you going to be able to build an application that just talks to codecs easily essentially. >> There's also like a wave of of um tools that like kind of work alongside you, right? Whether they watch your clicks or your screen and stuff like that. So you're saying basically that but for the agent. So like the agent needs to have like maximal contact with what you are doing and what you're seeing. >> Yes. Yes. the agent is always there is proactive knows where you're clicking around knows where you're on the website knows all the things what's also funny if you think about like chachi Atlas right so Atlas was kind of like this in a way right where you had like the browser and the agent the problem with Atlas was that Atlas was browser first and agent second this is agent first and browser second and because you do that it allows you to build a much more solid experience when it comes to like actually the building of the agent, right? And the rest is just like, you know, the browser just adds on at that point. And I think the reason why Atlas didn't succeed is because people are like, well, if I had to use a browser, I just use a browser. Like like it's it's faster for me to go to Google or, you know, go to magipad.ai then just ask codeex to go to magazi, right? Sorry, Atlas. But now because codeex it's the best agent and it's agent first I'm okay to having it interacted with the browser when I have to do stuff in the browser essentially. So yeah, basically kind of seeing in the future that like more and more of the tasks and more more of the work will be done by agents. Humans will then be providing the initial like spark some of the taste and you know so so then the alpha is okay what do the agents need? >> What do what does the interface look like and the I guess the easiest way to like stream the tokens both to the human and from the human. >> Yes, exactly. So actually a a really good point about this too that that you that you're making is like when you've built a product right now you have to think about the future of work is going to be less about doing the thing but more about supervising the thing. I believe that as the model gets better because they are getting better we've seen in the past you know two three years the model just keep getting better. the reason or the chances for humans to just go and say, "Oh, I want to change this pixel. I want to move this way." It's basically going to go to zero at some point, right? And so what you want to actually provide is the platform for you to supervise the agent versus doing the thing. What's very interesting about this, I want to show you um I want to I want to see if I can show you this as well. So here's Codeex on my app. Okay, this is this is codeex on my phone. So check this out. Imagine that this was in my hand, but I'm showing here on the screen so people can see. >> So look what I can see right now. I can say, "Hey, make a website in Magipath about David Andre and uh show it to me." So imagine I'm like walking in the street, right? Yeah. >> So because Magipath now it's a place for agent, >> it's in the cloud. Any agent can access to it, right? And so now it's going to do but because this is codeex is going to like you know look up online. It's going to look up you online. he's going to create a new project and we're going to be able to see this thing, right? And so for me, this is the future of work where like I am on the walk and let's say like you and I are working on a website together. You send me a Slack message or you put like a linear ticket and you say like, "Hey, Pedro, you know, like I want to change this." Why do I have to go on the desk and open something, I can just look at my phone and I can say like, "Hey, pull up the linear ticket that David wrote and and fix this design." Right? The future is this. It's like I'm not going to go into the thing and change it. I can just ask the agent to change it. Boom. Oh, look. David Andrew website. I mean, dude, this is magic. And like you can imagine, I could have been on a walk like like I, you know, I'm showing this as a screen share, but like I could have, I could have walked here on, you know, like a coffee shop and I could make this and then when I come back home, this is in my canvas. >> Okay. So, let's think about the alpha. Is it like a setup, right? Like obviously we talked about like the future of business and stuff, but like let's talk about the future of different individuals, right? Everybody wants to get an unfair advantage. Everybody wants to be on the cutting edge. So what will differentiate you from somebody else? Just a better setup. >> It's how easy you solve in a certain issue, right? So for example, the easy the the issue that we're solving right now, it's like well I want to go from the shorter delta from idea to share to collaboration, right? And so the way that people should think about when they make a new company, a new product, a new idea is first of all, what is the product I'm solving? And then how can I solve that problem so the agents can take advantage of it? That's the most important thing. >> Okay. That's the hard part. You know, that's where like the billion dollar ideas are made. So I mean, okay, like how would you how would you think about this? Let's say you're not doing magic path, right? Let's say you have to start something different. >> Yeah. >> Today, how would you think about it? >> I will think about it as as less UI as possible. Everything is some sort of API call or service that the agent can access to and ultimately it's about the data that you want to provide and let the agent work with right like let's say you and I were starting like a legal company like a legal AI company. >> I would never do a like I would never do a product right now. I will never start like you know like an RV or Lorra the other company like where it's like a product that you have to go and upload your dog on like why you have to do that right what I will do I will build a really good RNS so like a really good context for the agent to understand how law works all the different like um you know inside of the law and like making sure like everything is correct and then having codeex accessing that knowledge right plus my you know my PDF anything I'm working on for example right now I'm working one of my employees from Ukraine and I'm I'm uh sponsoring his 01 visa, right? And so there's a lot of documents, right? I would love to have a product where I just like I take all my PDF and I just talk to Codeex, but he also has some really good legal legal understanding so that I can like work with that, right? Yeah. >> Um and so basically the the the the the three things you should care is what is the problem I'm trying to solve? what is the context that I'm providing onto the to the model either in the form of knowledge or documents and then what's the easiest way to serve that product which now I believe is you know inside of the codeex app or cursor or you know whatever you prefer basically because this works everywhere like if I open cursor it works with courses too >> so making it easy for agents to use >> dude look at this that's insane search you so now I have a website magipath right and I can just >> real data the metrics are correct. >> It's the image is correct. Like you looked it up online and I can also bring this to Figma. That's the crazy thing. I can just like copy this to Figma if I want now. Right. >> So you're you're not seeing yourself as competing with Figma. >> No, because the clipboard is is blocked on because I I have to enable on on this browser. But it I'm not competing with Figma because Figma to me is still in the like pixel phase moving, you know? It's like ideation first. >> Yeah. >> Yeah. It's no agent first and also their file system is so complicated. That's why Figma hasn't adopted any of this stuff because it's like it's really hard for agent to build a Figma file because it's a it's a it's a cluster [ __ ] It's just like like even like a a super simple sign in card. It's millions of line of codes in a Figma JSON versus like you know 20 line of codes in React or like HTML or whatever. >> So even like the code base should be like simple for AI agents optimized for agents. >> Exactly. So the other thing that we did is like this can also pull from your repository like I can open like a a repo and I can say like oh bring my rep into the canvas show me like two four three four different variation of this idea and try to explore and then boom bring it back to the canvas right because the agent I like code first which is really cool >> very nice you mentioned the biggest leverage is brand and community can you like expand that >> yeah so similar to you where you know you build your own this like huge persona on YouTube like I built like a a persona on Twitter and it's Like you know in the early days I was showing like ideas around like how to use AI, how to use like different coding agent. I built my own coding agent. I remember like one of our first uh podcast was about cloud engineer right because it was like one of the very popular early like terminal coding agent. And so I think like even before you start thinking about what is the product that you want to build it's important to start putting yourself out and be like okay this is what I stand for. The one thing that is interesting about me if you like observe me from outside is like I was from the early days I was like this is the future of design. Like I remember I had this tweet that was like Figma is dead. There was like millions of people saw it. It became like a viral topic in the US and people call me crazy at the time. This was like this was like a year and a half ago. People like attack me. I thought I was insane but I was right. Like I knew this was coming. Um and and so I think >> you you really need to stick to your truth basically and like okay this is what I believe is going to happen become the person that people associate when they think about that and then build that product around essentially and of course like you need good brand you need good design that's all thing that you know plays into it as well of course. >> So okay I'm going to ask a selfish question here like for me if you were launching something new you would only think about products aligned with my audience. >> Yes. You don't want to get out of your niche, right? Like like let's say let's say like you know I've been talking about AI and coding, AI and design, AI and prototyping for so long that I think if tomorrow comes out and I launch like uh I don't know like uh even even AI related let's say I launch like a model training company right where it's like you go and like you put your your your weights and like like people be like whoa you know like why will I work with this guy? there's already like you know the file AAI and um open router and like all of this like like why would I use this product because I'm not that's not what people associate with me they associate me with like this type of stuff basically. >> Yeah. So okay let's say somebody is like starting out they don't have a huge brand >> how would that change or would you still like advise them to like build something in this way? My advice in that case if we really starting from zero is still to at least show something, right? Let's say you have like like because at the end of the day like I think the the the problem that people run into sometimes is like they don't just try the thing, right? And I know it sounds like kind of cliche where it's just like oh you have to do the thing but honestly the best advice I can give to people is just like just put yourself out like you never know. Like I don't know if you saw it there was this tweet that went viral recently the guy Fartza that did like the voice activation thing. >> Yeah. I mean like you know that went like super viral right and it was a good demo but between a demo and a product there's like a a C in between like there's like people that you need to hire there's like marketing all this stuff but at least now he has a starting point where people can be interested to it right um and it's also interesting like how much better the models have gotten right because it's like it's so easy now to build the experiences like that because actually this is also an interesting thing to talk about so just just to show you this. Um I I want to share this because I think this is important for the conversation. So I actually reshare that tweet from Farsza where you know he's showing like you know it's like basically like the demo for people that haven't seen it. He's just talking to his computer essentially right >> and I was and and I retweeted I was like you won't believe how hard was to make those thing in 2024. So this is me in 2024 talking to cloud 3.5 sonnet and but to do this you will have to combine so many different things right I would have to combine cloud code with the uh the text to speech then the speech to text and at the time model couldn't even run tools the way they run it today so you have to build your own tooling around having the model doing this stuff right all this to say now the models are so good that if you have a good idea it's very easy to just do a demo demo like this in two second with a coding agent and then having people get excited about like like because you can you can clearly tell that he built this with like cloud code or codeex because he's running in his terminal doesn't have any applications right and so my advice for you is like do you have a good idea just ask codeex to make it make it a nice video like this you know share yourself like talking about the product you just built if it's interesting people notice it right like I don't even think it's because he has a large following he has like around like followers this guy but I think the demo is generally Impressive, right? He's like talking to an AI. He's opening a website. Like he's doing a bunch of stuff. And now if this guy wants to go and raise money, he can probably go and raise money for this. It's very easy, right? Because he has like three million people have seen this video. So the most important thing is just like you just start. >> Let's talk about the demo because you've made a lot of successful demos over the last years. What makes a great demo? >> That's a great question. Dude, you're so good asking question. That's like my why I really like talking to you. So two things. one is the tool they're using. So the alpha, the absolute alpha is the one tool you want to use is this one called screen studio. For people that you don't know about this, this is the tool that everybody uses. And so what this allow to do when you're recording a video and you're clicking around, it's actually zooming there for you. So you can like very easily, you know, show the different part of the the thing that you demo. And in fact, this guy was actually using that. You can tell he was using that. The second thing is you need to make a video is very short, right? Like no more than a minute long. If you're like trying to impress people, like a minute to a minute and 30, it's enough. And three, it's like just make a very coherent story, right? Like the reason why the Fars will work really well is because you know he's like open Spotify, ask you play a song, roll with the volume, and then he's ask you like, oh, can you show me like my revenue for my product, blah blah blah. So it's a it's a very compelling story of somebody using their computer. So find like a simple story to explain and then boom make that like so for example um you know recently one thing that uh blew up for me and and magipath was showing how magpipath works inside of codeex which is basically what I just show you right or cursor >> um and so when I did that that that sort of like uh announcement literally what I post was this and this was uh uh wait no this is not this the video one second. It was this one. So, this is a 43 second video, right? Almost half a million people saw it. And all I'm showing is I'm opening the browser. The I'm going to Magipath. I'm asking for four different music application. And then Codex builds the music application and they're actually interactive. I mean, the audio won't come up in this video, but like you can play music in this different like music player that he has built, right? 40 second using Screen Studio. very simple idea. Boom. You can use magic pivot as an external canvas into into course into codeex and you know half a million people most 400,000 people saw it and of course now that made our user base goes like that right because this is a something that people want people want a canvas for codeex. >> So basically also thinking about the distribution right like as people are building it how to demo it how to put it out there. >> Exactly. So like what advice do you have for people because a lot of them get stuck in the building stage forever and you know they focus on adding features and they kind of leave it to the sides the distribution. >> Yeah. So my advice is like perfect is the enemy of good. You kind of have to first solve one problem right when we first came up with magic path we have one feature which was just like AI generates design that's it right and and the way we used to talk about oursel was like the tool for AI to generate designs right and that was it. and we did that really well and and that allows it to grow but you can't have like a million feature because then it's really hard for people to understand what you're doing. Um the message get diluted and when it comes to distribution like you really need to understand who your user are. So for example before we launch Majipa 2.0 know I went to the in my discord and I asked 20 people there to be my beta tester and I because I sometimes like as a founder too or like when you when you build a product like you are so into it that you don't see the large >> right >> and so when you have people coming to you and say oh well actually I like this I don't like this this is good this is not good then it moves you to the right direction so you do need >> to put this stuff in front of people because unfortunately the problem is like not everybody's Steve Jobs right like like what made his job special was that he could just know what people wanted before having people even to see it. But that's like a godlike ability, right? And and it's and you only acquire after building product for like, you know, 30 40 years that he was doing it. Um, and so you kind of have to have that constant messaging with the people that you think might use your product. So touch with reality basically you know like you can easily deceive yourself that your thing is great and you know you as the builder the founder you spend the most time with it right so you know all the features you know the user interface all that stuff and if you're the only user or maybe couple of your friends you will never see the rough edges >> exactly and and the other thing too AI is so exciting right now that people would just literally do that for free like the way that I do is like I offer magic path credits right but like if you're not like if you're a new founder and you say like, "Hey, I have a good AI idea. I'm looking for tester." You will get so many response online because AI is so hot right now and it's been hot for a while and it's going to going to get hotter and that's just what people want. Like if you say like, "Hey guys, I'm testing a new uh AI tool for riding, you will get an insane response, right? Um and and and because it's such a like exciting time right now. So you you can find tester so easily. You just have to tweet. You just have to tweet like what you're doing basically. So Twitter is basically the place to go, you're saying, >> for people to both learn about AI and demo it and promote it. >> Exactly. I think like Twitter is it's still number one in terms of like alpha like things are happening where people are moving like like I knew that people were moving were jumping ship from cloud code to cursor to codeex like so early on because I I I just see it. I was like interesting like a lot of people are talking about codeex now like and now you can see it's just like it's it's almost flipping right. Um, and so I think Twitter still remains the best place for alpha, but then to learn, I would say that YouTube is still probably the best. and and actually YouTube is the one that probably gets you more customer because yes, uh Twitter X, whatever you want to call it, as the most kind of like alpha uh people, but it's also like a very small sort of like uh bubble of people that like AI obsessive. But like if you actually want to break into the market and you want a product successful, you have to put on YouTube because that's where like you know the average your average customer basically lives. >> Yeah. Yeah. And also like people watch like I can see it from the retention charts, right? Like the same video I upload on Twitter and YouTube, even if it does well on Twitter, people just drop off after like 30 seconds, 60 seconds. On YouTube, the watch times are like way way longer, like seven times longer >> because it's just not a platform to make to watch video, which is why like I try to stay like under a minute when I post something on Twitter because I know their attention span of people is just like so low um that I'm just not I'm not going to do it. >> Yeah. So basically, you think everybody building right now should be willing to just get on camera and just demo their thing because like a lot of people I'm talking to, they're kind of scared of this. You know, there's they're not they think they're not social media per people and all that stuff. You think those just limiting beliefs? >> Yeah, I think they're just limiting themselves. Like it really doesn't matter. Like you know how many apps I I made in in Codex that like I could just like sell if I wanted. And the reason why I don't do it is because like Magic Puff is my is my main uh like interest right now. But like I made so many apps are like actually really cool like like I can show you this one actually. It's really interesting. It's called Audi sale. So what this does is so I can I can write right? It says like hello I'm recording a podcast and then I can hit tab and it like autocomplete like what I want, >> right? Or like I can also speak like I can say like um so it's almost like a mindto text connection. So check this like Google Docs would be. >> Yeah. Right. So, it's like, oh, I'm here. I'm here. You know, I'm I'm with with David. We're recording uh a podcast right now. And then I can like tab autocomplete. Like, how insane is that, right? That you can just like do that, right? Or like imagine like I can also say let's say like I want to write something in between here. I can put a parenthesis and then I can autocomplete even complete in between sentences, right? So like imagine like if I post this online it'll probably go like so viral because such a no nobody has done of writing uh things like this right and like there's so many ideas there's so many opportunities around AI that like nobody's using especially with like stuff like GPT real time and text to speech like it's just like an entire new category of product >> and you just need to find the thing that you're passionate about and just like post it and people will eventually they would they would buy it. >> Yeah. Um and you mentioned something great is like there's a lot of you know SDKs there's a lot of AI tools that people aren't using like everybody looks at you know Opus everybody looks at GBD 5.5 and you think about like LLM based things which again like that has been around for you know four years now really working well that wave of like startups you can see like perplexity uh ker are like the biggest ones but people don't even look at like you know what they could build with nanobatu what they could build with you know the GBD real time like you said there's like going to be many of companies that just figure out one use case and either get acquired or build a large company. >> Yeah, exactly. And just like and just like run that user case, right? And then because like it goes back to the whole idea of community and brand, right? Like once you get that winning thing, it doesn't matter if you're not OpenAI or entropic like if you get like you know like 5 millionR like everybody wants to buy you, right? Um and so you know we we you won't believe at the amount of like buying conversation that we had at Magic Path, right? But of course I want to grow this company as much as I can. But the the thing is like yeah like you you can just find that niche and just run it and then eventually like if you do it right you can succeed because at the end of the day AI is just one thing one thing I want people to understand that to make the best product in AI it's just a context problem. you provide the right context, this thing works. When people say like AI is slop is because they don't provide the right context. Like you know how many people hate 5.5 at design? Like look what we just did. Like this doesn't even look like 5.5. This look like hoppers, right? But it's just because I provide the right content. Like the interesting thing too here it generated an image, right? He went and made his own image here to use like GPT generation image to create to put it into the website. Like I don't even have to say it to do that, right? I mean those models are just so smart right now. All it matters is the context. >> Okay. You mentioned, you know, the the current stage of the startup. You are, you know, you're getting offers to get acquired. How was the different stages? Like you mentioned it's like around one year old, a little bit over. Explain to like the different problems at each stage. What people need to get right on each stage. >> Yeah. So like basically the life of a founder, you are alternating between the best day of your life and the worst day of your life constantly. Right. So, Magik is actually an interesting example because when we launched, we were the only product that was doing AI and design and I was called crazy for it, right? As I said before when I was talking about Figma. >> Now, it's like everybody's [ __ ] doing this thing. Like everybody, it's like you get, you know, you get uh uh uh claw design, which I think it's crazy. It's like an awful product. Nobody's using it, right? But you know the normie on the street that never heard about AI and design here it's cloud and design of course he's going to use it which is very unfortunate because you get this like when a company that big try to take a name like design and make it their thing it's it's not as good for the startup ecosystem right because you know we're all trying to compete for attention and and and distribution all that and so I think like the one thing that's important is that at any stage of a startup you always going to compete for attention and you're always going to have competition. The only thing that matters is that you stick to your to your ideas and like you stick to what makes you unique. For example, when I was saying like I don't see myself competing with Figma is that in Magic Path for instance like we offer some like edits, right? And like you can change text, you change the color, you can put the gradient and border like all the stuff, right? The thing you cannot do it's like moving stuff around and that's because like this thing is deployed like this is like a real react website. It's not like, you know, your Figma JSON file, right? And so there's an argument you could be making. It's like, well, why don't you do that, P? Like, why don't you make this editor better? Why don't you invent a new technology that makes the React editing? But the thing is like, I don't I don't want to compete with that because Figma spent the past 15 years building the best editing experience for any kind of this type of experience. So, it's like I can only lean into the things that make me unique, which is it's fully agentic. It's uh something that AI really understands. and you know anything is like deployed and and lives in in in the browser like I can just like visit any website that I create right that's what makes us special and so I think like as a founder as you get to this like competition any stage and and and and just like you know all these different competitor that you have you can just stick for what made you special in the first place and you just like keep keep running that basically >> okay so why don't like you mentioned you want to build this as big as possible how did you make that decision how would you like advise people what to build for, right? Because there's two arguments. Obviously, you know, go as big as possible, as ambitious as possible. But the counterpoint to that is like the future is very uncertain, right? Like people can maybe try to see three to six months ahead of time. We don't know what things will look like past AGI or super intelligence. I don't know if you believe in like short short timelines, long timelines, but like how would you advise to people like okay, build something that people know right now is the need for it. Right now is the time for it. maybe you know go for a quick acquisition or try to like predict where the future will be going build something that the agents will use. Yeah. How would you think about that? >> Yeah, that's a great question. So the basically the only thing you can do is taking one or two bet right and one or one or two things that you are like so sure is going to happen because as you say things change so much I mean remember like when we first talk about coding agent now it's just like you know any like four months ago nobody will even thought that anthropic was going to be dethrone as a coding agent now everybody's on codeex right so thing changed so so much so it's very tough to predict necessarily what people are going to need. The only thing you can bet is the models are getting better. The models are constantly getting better and whatever you want to call it like super intelligence, post agi, whatever that's going to happen up to that point, the model is just constantly going to get better. So if that's the thing you believe and I think that's the only thing you can believe, that's what you build around it, right? And so for instance like you know even when I was working on entropic and I was working on like on search and cloud code and MCP all those stuff I think I stood out inside of the company because I was building almost for like you know oppos 7 like I was building the tools and application thinking that those models were already intelligence and and so what what that what's very interesting about that is that how you build those application and experience you start to see that the models are not there quite yet but they are good at doing certain things Right? And so the only thing you can do right now is saying well I believe this is the future of design in my case and I know that's where we're going to go because the models are just going to get better and I can only build toward their future incrementally right so magipath initially was a tool where you can design now it's a platform for aging and I know that you know six months from now this conversation will be very different where instead of me asking codeex what to do what probably is going to happen is if you and I were repeating this in six months from now you and I are talking and Codex is just making stuff for us. It's a proactive agent, right? It's it's not the agent that you ask a question. It's the agent that understands what's happening and work for you, right? >> And so that's the only bat you can take is AI is getting better. Pick up a a subset of human experience that you think AI is going to be amazing at and just build toward that essentially. >> Yeah, I I agree completely. Can you show us like how you actually build software, right? Like do you use the Codex app? Show >> us. So for for me, you know, so everything that we're doing designwise, every design thing is built in Magikipath and then we use like, you know, whatever we build in Magikipath, we send it back to the agent to implement into our repository. But generally when I like if I want to start something new or I want to try a new idea, right? Let's say like I'm I'm in codeex. I have all those like quick uh text replacement in my keyboard. So like there's things that I can do. So for example, if I if I say absorb, it just writes this whole thing which is like absorb this code deeply uh you know sort all the feature in the code. So that I start that say I'm starting with a new repository. I select that and I start or then I have this one is my favorite which I use all the time which is >> which tool is that? This is uh I'm asking basically to uh explain this thing you're working on as you would explain for a junior developer and then write a plan for the junior developer to build that. And the reason why I like >> is because if I use plan mode in in codeex let's say I was using like I think it's like slash plan I think it is. Um, so if I do this and then I want to change the model after the plan, it doesn't allow me to do that because if you select, let's say you select like 5.5 high in plan mode that you can change to low mode to implement that plan, right? Because the thing is like the the smartest model you need is the one that needs to figure out the problem. >> Then you can just use a smaller model to fix it. So the way I like this is because if I ask you know 5.5 high to come up with a plan for a junior developer then I go here I change it to low and then I get that one actually implementing the plan and then another one that I have that I love I say uh if I spelled this I basically told it to behave as an agent. So I was like, "Okay, remember you're in an agent. You need to compose the query blah blah blah blah blah." You do all of those things. And what's funny, this is basically goal mode, right? Like so today you don't even I don't I don't I use this rarely like you can just do like slash goal, >> right? And it just you can just write your goal. >> Um another one that I love if I do spawn >> it asks to spawn multiple. So actually this is like really cool, right? Because let's say I say make a few images using uh your image skills of I don't know electric Ferrari since this was like everybody was talking about it. >> Yeah. >> Um so if I keep my spawn multiple agent um uh prompt right now it's actually going to do that. It's going to ask multiple agent to make images. And you can imagine you can do the same thing with your coding. you can say like oh I'm building this thing spawn multiple agents right um and and what's awesome about this is like you build the thing in like five minutes instead of like you know an hour or whatever like whatever it will take you to build that thing so I basically have all these commands right or like for example this is my favorite tool if I if I write PR it says push a PR towards that branch uh you know with the things I'm working on or this one is even crazier so for example a lot of times my my designer my sorry my developer operate will work on a branch and then I want to take that branch and make some changes and maybe push it into main right which is like the the main product but so what I have here it's like this prompt what this does is say okay take this branch that I'm working on make sure you pull from the main most recent uh branch check all everything that matches cherry pick the things that um you know are working and then implement that so it's like it's such a it's almost like my brain in a way is connected with with um with codec because it's just like I have all those quick things that I can write or if I if I write like explain it >> you have this like in a GitHub repo like this would go viral for sure. >> No, I I know I should probably do this. No, it's just it's just text replacement. I I just use like keyboard actually. You know what's crazy? You can use codeex to build this text replacement for you. Like that's what I do too. I ask codeex like hey go in my text replacement. Can you add me a few new ones? And it just does it for you which is kind of crazy. Yeah, I think you should just like take 10 minutes, throw this in a GitHub repo, name it pro skills or whatever and just, you know, release it to be fair because like there's so many GitHub repos that just like get like 30k 40k stars in a less than a month. That's like >> Yeah, that's just those people's workflows, right? Like it for us maybe it doesn't seem as valuable because like we do this [ __ ] every day, but for people who don't, this is the gold. >> This is huge. Yeah, this has changed my life. Like I I'm so fast at developing and and and designing all this stuff cuz it's like, wow, this looks so much better than the >> Yeah, it's crazy. A [ __ ] AI model is is able to do better design work. >> Like that's crazy. Like that's the electric Ferrari I want. >> This is 600,000, you know? Not not that [ __ ] >> Yeah, the the Lucha [ __ ] sucks, dude. Like this is this is so much better. Oh, because I'm zoom out, that's why. Oh my god, dude. So funny. Uh but yeah, no, it's a it's it's it's super cool. I do also have a lot of skills like like like that's the thing like you don't even need skills when you have stuff like this like I can say like debug and you write these like really good things like review the recent code that we've wrote look for potential bug regression edge case side effects and then list all the risk like like why would I use skills like I like this is basically what that's doing right and like you and of course like you can write like some really long one like >> but you can do like slash right like with skills you can invoke it with slash >> yeah exactly but for me it's like you know it's so much faster to just have those like keyboard because then the cool thing if you're using a MacBook and an iPhone then I also have those thing on my iPhone. So if the same you know it's it's saved on my iPhone too. So when I'm on code that's mobile I can do the same thing right. So I don't need to have I mean skills I guess are connected but um yeah it's just it's just a better way basically to to to work. Yeah. >> So basically anything you find yourself repeating say like what more than once a day you turn it into these keyword replacements. So the way you do it, you go to settings um and then I do like yeah text replacement let's call um and then they're all here. See, >> but truly you can say you cannot you can ask codeex to make new ones like like look I can literally say like this is crazy like like look in my text replacements for the keyboard and suggest a few new ones. Yeah, this is like a lot of alpha as well like realizing how how much control it has over your computer. Like for example now I don't even use folders. If I'm like doing something in a markdown file, say like open this in Finder for me and instead of like clicking seven folders deep. >> Yes. Yeah. I mean that's literally like the the the the future, right? This is like this is the best computer. Like look at this. You come up with like oh I like Andoff, right? Like okay, okay, cool. Add right now. It's going to add that in my text replacement. I mean that's [ __ ] insane, right? Like it's like >> it's crazy, right? >> Yeah. Also like people people don't know they see like okay cloth code or codeex right and instantly think about coding but like it can do anything in the computer right like organizing your files finding the largest files optimizing this place one of my friends was had like an old computer from his dad like 15 years old it was lagging and [ __ ] he just used I don't know cloud code or codex said like make this faster it ran for like 5 10 minutes you know deleted some files some apps optimized processes and it was like way faster than before >> no it's insane yeah you can fix your computer you can fix your Wi-Fi connection It can download anything, dude. It's in Actually, let me show you this. This is really cool. Give me a second. I want to show this thing. So, I have this device here. It's called a flipper. Although the battery is dead. Let me This is what I was going to show you. This is actually really cool. So, here I have it's this is called um a flipper. So, this is basically like a way to do security control on network. You can also like create like uh uh cards like you can like, you know, duplicate cards. And I was like, you know, like I don't know the coding language of this thing. It's a device. Like, what the [ __ ] do I know, right? Yeah. So, I connected to Codeex and I was just like, "Can you just make a game for this thing, dude?" It made me like this like double like like doodle jump. I don't know if you can see. And it's like, see, I'm like, I can actually play this game, right? But the crazy thing, it made the screen vertical. It was because it probably Okay. Well, control like you figure out or like I can use as I can also use as a music player now or the all the interesting thing that I build when Codex send me a message this thing says that vibrate so I know that Codex has finished to to send a message which is like how cool like like I I don't know how to develop for external device or like or like he hacked this like um little e-reader so he can make now he made me an app for for this thing like it's it's insane we're we're really like in the in most insane era of technology and humanity. It's just like so exciting. >> So basically acquiring as much hardware as possible, improving your AI setup as much as possible so you can smoothly turn your ideas into reality. >> Yes, exactly. There's so much such a such a fun time to act and like you know do new things is like just like amazing. Yeah. >> Okay, let's leave people off with a clear action step. If you were somebody who just like has, you know, the default codeex or default cloth code, >> what's like one or two things you would do today to improve your setup? >> Yes. So, first of all, get uh any skills that makes you do what I basically show you like just like improve your workflow. At the end of the day, engineering is a problem of context, coding, debugging, right? So what you want to do you want to get very sharp in the way that you provide context to the AI but then also how's the AI is building the code and then also how you looking for bugs in the code right so you have you have to have a really solid test sort of pipeline that would be the the the first thing second thing if you need like a a canvas to you know design and build actual functional apps of course I'm biased here but Magipath is amazing um I didn't show you how to install it but it's basically a skills it's a maybe we can we added to the YouTube description, but you just install the magic path skills and then all those agent can >> or just tell your agent to browse the web and install Magic Path. >> Yeah. Yeah. Exactly. But it need it needs the skill basically, right? To to know how to communicate with Magic Path. Um and then the third thing would be like yeah, just build as many ideas as possible. Uh don't be shy like nobody's going to judge you if you get three likes on Twitter. Like that's I I I like in the early days I will get two likes on Twitter like like it happens like when you start it's like normal and then you know eventually you build something that people are going to be like oh okay this is cool and then potentially like happened to me that's how you raise money like I left on tropic I built a demo magipath I shared a tweet I think I got like a million views on that tweet that was like here's how you design in the age of AI and then I raised money in a week after that. So, uh, >> that's like the TLDDR, you know. Like, >> of course, like I I I I, you know, I'm very grateful. I know I was at the right time in the right place. >> Yeah. Lots of work, great taste, you know, experimenting. >> Exactly. >> Yeah. There's like years of work. It's not it happened overnight, but at the same time, you kind of, if I didn't put that out, I wouldn't do this, you know. So, yeah. >> Yeah. >> Right, man. Appreciate for your time. Lots of amazing advice. I took [ __ ] ton of notes myself and u hopefully next time it's not a year until I have you on. >> Yeah. Yeah. Yeah. Yeah. We'll we'll we'll chat more once the new OpenAI model comes up. >> All right, man. Maybe there's a hint that it might be sooner than people think. All right, Petro, have a great day.