Transcription
Everyone's talking about Manis AI, the Chinese AI app that basically can take your thoughts, turn your ideas into fully automated businesses and products. Now I wanted to try this, but I didn't have access, so I called my friend Min Choy, who came on the podcast, and we had a crazy idea: can we build a business—actually, the idea was a DocuSign killer, multi-billion dollar business—in some, a few prompts? In this episode, we go through it, and we show you the ups, we show you the downs, we show you what's good about Manis, we show you what's bad about Manis. But if you're here, if you care about AI, if you care about building businesses, understanding how Manis works is a must. I was so impressed with some of the stuff it was doing, and I was so not impressed with some of the other stuff it was doing. Uh, this video shows you all about that: the good, the bad, the ugly. I recommend you watch the entire episode, uh, and you make the judgment call for if you can go and create a business fully automated with a few prompts on Manis AI that generates $1,000 per day. Enjoy the episode, time baby! All right, I brought Min Choy today. Min has been going viral recently because he's been playing around with Manis AI, and he's going to give us a first look into how to use Manis to actually build a business that works for you, Min. Thanks for taking the time on coming on the show. What is the viewer, the listener, what are they going to get out of this episode by listening all the way through?
So thanks for having me, Greg. Uh, this is a very, uh, exciting podcast to join. Uh, it's my first time, so thank you. Um, so this Manis has been going viral and very crazy in China, and it just like came over like just a few days ago, and I got an invite, um, so that I can go play with it, and I realized that this thing is just more than the typical AI that we've been seeing. And the thing that is different with this is that we've had, in the past, uh, that we wanted to make this agent AI where the AI can actually go and like run tasks and run tools, um, and as you've seen on like Cloud computer use and browser use and similar to that, um, and this is sort of like more advanced, uh, the next stage of that. I think that we're getting closer and closer to sort of the glimpse into what the AGI is going to look like. So, um, what I want to show you today is that I, I want to take some of you, some of your, um, startup ideas, correct, and I want to see how far Manis, uh, AI can take it. And we're going to try to do this, you know, step by step and, uh, in a very, uh, like a process-oriented way, um, and we want to see if this thing can really like help entrepreneurs or startup, uh, owners to really kickstart their ideas into action, um, and I think that, uh, that this will show, uh, like what this type of technology can do for us, uh, in terms of growing a business. Okay, let's, uh, let's see. I mean, I'm curious. This is my first time getting a glimpse of Manis. Um, I hear I hear big things, so yeah, let's just let's get right into it.
Yeah, let's, uh, start with some of your great ideas. Um, I think that one of the things that I, uh, read recently from your post, uh, was something about DocuSign. I think DocuSign might be a little bit complex, but let's give it a shot and see what Manis can do for us. Okay, and while you pull that up, my thought on DocuSign, Min, was basically that, um, what's happening is, you know, DocuSign makes, I think, like billions of dollars or something, thousands of employees, um, for an electronic signature, and with AI, it's going to just be a lot easier to create software—we'll see by the end of this episode—to compete with, you know, the DocuSigns of the world. And instead of charging $10 a seat or whatever they charge, you'll be able to make it free or charge just a very small amount. So I think that there's an opportunity to build—I mean, I think multiple people could do this—can make about, you know, at least $1,000 a day in revenue, um, and so I'm excited to see if if this could work. All right, so what do we—it says hello, Min, what can I do for you? Um, yeah, so this is this is a very, uh, familiar interface that we see all time, uh, you know, ChatGPT kind of had this type of interface, and everyone else kind of start copying that, um, interface design. So Manis, uh, also has a similar, you know, you got a little chat window with the you speak with the agent, uh, so you just kind of simply say what you want, um, and I, I, I like to start very, uh, simple, uh, request, uh, like, you know, let's say, um, create a DocuSign clone.
Quick break in the Pod to tell you a little bit about Startup Empire. So Startup Empire is my private membership where it's a bunch of people like me, like you, who want to build out their startup ideas. Now they're looking for content to help accelerate that; they're looking for potential co-founders; they're looking for, uh, tutorials from, um, people like me to come in and tell them how do you do email marketing, how do you build an audience, how do you go viral on Twitter, all these different things. That's exactly what Startup Empire is, and it's for people who want to start a startup but are looking for ideas, or it's for people who have a startup, but just they're not seeing the traction, uh, that they need. So you can check out the link to StartupEmpire.co in the description. Okay, so simple, like a super simple request. Very simple. Yeah, I start by there. Um, the cool thing about this is that Manis is sort of a multi-agent, um, under the hood, and it's got like planner, executor, that type of thing. And if you remember like very early in the days, uh, when we used to, uh, have this, uh, thing called BabyAGI, uh, by Yo, I don't know if you remember, um, that kind of used to have that sort of that setup where it had like task creator and like an executor and etc. So this thing does a similar thing, and then what it's going to do is—GNA—it's going to take your request and then it's going to try to research into what that is and what it needs to do to create it. And so even giving a simple request, um, is doing many things, uh, like on its own: like it's planning, it's designing, it's executing, and it's going to try to run things and then analyze the results and then try to kind of replan or update as needed if it's doing, if it's going the right way. So you will see a lot of things, lot—there's a lot going on—it's doing a lot of planning that I'm researching right now, so it's, uh, doing it, you know, sort of the Googling right now under the hood, and it also checks out their website, uh, I think it takes—Al—it also takes screenshots, um, and it tries to do stuff on it. Um, I did like run into one, one interesting thing before was that it, it, it got stuck on like captchas and stuff, and I thought that was kind of funny, and it wanted me to like kind of intervene and like do something with it. So it, it has some sort of the, uh, human-in-the-loop, um, feature in there as well, um, so it will ask you for confirmation or, you know, additional contacts as needed. So this is, so this is interesting. So if I'm understanding this correctly, you know, before Manis, you kind of had to go to a bunch of different places—I call it the AI cocktail approach—so you would go to maybe a ChatGPT, um, or a Grok3 to kind of come up with the idea and create a PRD, and then you'd go to Figma to actually design what it would look like, and then you'd go to, you know, a Bubble, a Webflow, a Cursor, depending where you are and, you know, in terms of technical prowess, actually go and build the idea, um, and deploy it. What, what you're saying here is what's happening on the screen, um, is it's kind of doing all those things at the same time.
Exactly. Yeah, it's, it's actually what it's doing is it creates a sandbox, um, like a virtual sandbox, uh, of its own, and it, it runs like, you know, command-line, uh, executions, uh, it runs a browser, runs tools, uh, which is on the right; it's right now is browsing and just capturing stuff, capturing information. I think it's what, what it's doing is it's, is crawling it; um, it's, it's scraping all the data, uh, and information it can get, uh, so since I said DocuSign, goes to DocuSign, and it checks out like what, you know, type of information and the feature, the, the product has, and then it will probably test it out, uh, and see like what feature that can capture, uh, as part of its planning. So, and remember like we, we just started with this very simple to say just go create me a DocuSign clone, and it's now just going there and doing its own thing, and it's just pretty, pretty wild to see that, uh, where we are with this agentic AI that is able to, um, kind of, uh, do most of this task, uh, on its own, understands your request, and if it doesn't have all the information, it goes and research on its own, and I think that's like the key to a lot of this, and like you said that we've had many AI tools, uh, that we had to use like separately, and we've done a lot of copy-pasting in between, um, now a lot of that has improved a lot with Undercursor, Webflow, uh, you know, Replit, and all that, but this is a little bit different because it does more than just coding; it actually does the research; it actually goes to the browser, and it actually catches, uh, captures the, uh, information from there directly, and, and then it takes all that context into its planning, uh, and then it, it formulates a plan of execution, and then it goes and executes them. So this, it's a very interesting way. Now you, what you see on the right side of the markup, um, the, the way it has that markup, it, this reminds me like a lot from like the BabyAGI days. BabyAGI used to kind of do this in the command line where it shows like it's planning phase and all that stuff, and I thought this was interesting. So it's doing like step by step on what, what it needs to do to create this DocuSign clone, and that's all I said: just go get create me a DocuSign clone, and it's, it's getting all of that information, and it's planning, replanning, and, um, executing. So it's, it's very interesting to see, and it takes—it's going to take a couple of minutes, um, to get to the end, um, and also sometimes as task as it's needed, uh, it may have missed something like in the process of doing this, so, um, yeah, there's going to be a lot going on on the, on the, on the screen now, but it's, it's going to be pretty interesting what it comes out with.
So have you ever had an instance where you're like, create this app, and you're looking at maybe like the markdown, like for example, if you, you know, if you go back to the markdown, can you go back to the markdown while it's, or, or it's, um, yeah, I mean, you could like look at the tools. Oh, wow, wow, that, that markdown was just like a—it's actually the raw source code, but this is the, uh, preview. So cool. So, so I'm, I'm actually going to read a little bit of this. We do have a lot of listeners on Spotify and Apple for the, uh, Startup Ads podcast, um, so it says here: research DocuSign core features, identify key functionality to implement docu, uh, document user flows and technical requirements. Now there could be a case where you're like, whoa, whoa, you know, I don't want you to do XYZ; I don't want you to identify the key functionalities; I just want you to, you know, maybe only focus on this one feature, um, is there—have you ever asked Manis to do something and then kind of said like, oh, I made a mistake, or you did something wrong? Is that—yeah, you can actually, uh, talk to it in this middle of its progress, so, um, which key function, like, do you want to limit this to? Well, that's a great question. I'm going to pull up the, the DocuSign website right now, um, cuz it looks like when I'm on the DocuSign website, it looks like there's—they've created a lot of different products at this point, so like, for example, they have the e-signature product, the payments product—they identify—all I care about is the e-signature product, so I just want to confirm that we're only talking about the e-signature product. So let's only focus on the e-signature key functionalities. Good. Yes, that's cool. Well, that's a really—that's actually a really helpful feature, cuz like if you're using ChatGPT, for example, it's, it's in thinking mode, and when it's in thinking mode, you're kind of twirling your thumbs unless you stop it and then, yeah, ask it to do something. Yeah, this is, this is probably the one of the coolest features that I've noticed on this particular product is that I can interrupt it in between without it like, you know, stopping its work that it's going to try to—as you see here, it's going to say, I understand you want to focus specifically on the e-signature key functionalities, and it will adjust the approach to prioritize the core e-signature feature of DocuSign. So, and then it says, let me continue in setting up the project structure with this focus in mind. So that's—this is like one of, one of the coolest features that I've seen on this—that you can actually kind of, uh, as sort of a human in the loop, like interrupt the process to, uh, readjust, uh, the planning or the setup. That's cool. I also just noticed that like it's very clean; like I, I feel, I feel like I understand what's going on; like I, it's very pleasant to the eyes.
Yeah, a lot of the, a lot of the demos that I've seen, uh, most people just kind of kick this off and just kind of walk away and then just, uh, you know, have a coffee or a snack or a lunch and just come back and just check like, you know, where it's at. So it's kind of funny, and, uh, I, I know you've seen those videos like with the guy with the person that's on a Tesla driving, and, you know, the person is on his way to some coffee meeting, like, oh, you know, create, create me some, you know, talking points for this coffee meeting, and it's just kind of driving—zone—and the Tesla, FSC is driving him, and also, uh, Manis AI is preparing his, you know, meeting notes. It's, it's kind of funny. We'll, we'll include that video so people could see that what's—we, we'll show that video on screen, so yeah, yeah, yeah, yeah, yeah, yeah. I don't, I don't like that video that much because, uh, it, it just feels like I don't know what's going on other than I see a car and then screen—the screen's kind of small—so I don't, I don't know if there's doing anything, but I thought it was kind of cool, just the fact that he was doing it. I thought it was cool. How long is the, you know, how long do you anticipate this taking? Is it like going to be 10, 15 minutes, 20 minutes? Um, depends, um, you know, if it's something very simple, like, lot—some of the research stuff, it actually doesn't take that long, um, so I, I've done a, uh, research, uh, to say, hey, buy me all the research, uh, on this particular topic, and it'll go and just, uh, do does this Googling and searching, and then it'll find it and then just kind of bring the report, just like very similar to, you know, the Deep Research and the Deep Search, etc. So while this is running, you know, what we can do is, uh, start a new session and say, let's get like top, um, Greg's startup ideas and see which is the most viable to run on Manis while, while, while it's doing the—how's that? Beautiful. Yeah, I think, I think for me, it's, it's, it's most viable to run on Manis, but also the one with the—what does Manis think has the highest probability of success? So I'm interested to know—I'm interested to see like what is—how intelligent is Manis, right? Like there's—on this show, on the Startup Ideas podcast, I've given away, I don't know, a thousand-plus ideas, and the reality is some ideas aren't that good, some ideas are good, and some ideas are really good.
Yep, yep. And it's, it's really—it's really the listener today, it's the, the humans, the listeners who have to figure out, okay, this is a really good idea, or, you know, what, Greg had too much wine, and I don't know what he was thinking when he came, when he shared that idea. You know, you got some really great, great ones. I, I'm surprised that people haven't taken some of those ideas and put it into—you know, put those ideas to work. You know, all this AI tools, they do, but a lot of them just DM—like they're, they're like, they keep it to themselves, and they just DM me and they're like, I built this thing, and it's doing like—yeah, it's crushing. I guess if this was, you know, so easy, then everyone will be all like rich, right? All right, so I asked, uh, I asked Manis while is doing his DocuSign, um, creation to research into your startup ideas and the top, top ideas that Manis thinks has the highest potential—that's what it comes up with—and see if we can have some actionable, uh, you know, startup ideas that we can also ask you to create while it's doing the DocuSign creation. Okay, I have an, an edit to this prompt, sorry to do this to you, Min, okay, but, uh, I would say I think we should ask for ideas specifically that I've shared on X and ideas that I've shared on the Startup Ideas podcast and the Greg Eisenberg YouTube channel. I think we should just, uh, qualify that we, that's where we want—or what do you think? Should we, should we qualify it, or should we just let Manis figure it out? Let's, uh, let, let Manis figure it out. All right. Yeah, let's—the thing with the, the thing with this type of agentic AI that I like is that, um, the—it's easier to start with, uh, that—if you notice like on the other AI tools, uh, people's, people are running into issues because they don't know where to start; they don't know how to ask; and, you know, they, they're not maybe not as good at prompting or have templates and etc. to, you know, get their ideas across, but you click the browser, by the way, sorry, could you click the browser? Like, can it go more full screen? Okay, wow. So this is—it actually—so this is—it's on our, it's on my holding company's website right now, which is, by the way, insane. I'm surprised—checking out your—it's checking you out, so oh my God, it's, it's going to, it's going to kind of crawl through your site and just check out what you got and, you know, grab as much of the information as possible. I'm just surprised, honestly, that it knows how to navigate the website because it's like we created a world, so it's kind of like confusing.
Yeah, it's, uh, it's using a, uh, tool, um, called, uh, I believe it's Puppeteer, uh, which is like an automation tool, so it's a, it's a, it automates browser, uh, tasks; it's a, uh, it's a tool that's been around for a while in test automation world, um, so, and it's using that to, uh, do a lot of these things. It's similar to like what Browser Use and, uh, you know, Cloud Computer Use, uh, uses. Interesting. Yeah, see all I, you know, I, I just have your name and your just a very broad ask, and it just goes into this research mode to kind of find out about you, that now it's going to—going to your Twitter—that's crazy. But it doesn't have—say Twitter X—right? It doesn't have an X account though, right? So it can only see a couple, a few of my tweets. Oh, wow. So it knew to click because in my, in my bio, it says—bio—it says, click for free startup ideas; it knew to click that. Wow, that's crazy. Yeah, it's, uh, it's, it's a different type of agent, agentic AI, I would say, and I know you—I know there has been a lot of things going on, uh, just like the other day, uh, that people found out, uh, that, you know, Manis AI is really under the hood running like, Claude. Yeah, yeah, there was a lot of memes going around talking about how, yeah, this is just a wrapper, and etc., that may be—and then may be very oversimplifying it, uh, I, I think, uh, what they've done is create like a whole framework under the hood. I believe is that they got a, you know, army of agents under the hood with many different fine-tuned, uh, AI models that it's using for specific tasks, that—and that, that's why it's doing what it's doing, and I think it's—they're, they're—they've done a really good job, in my opinion, uh, in terms of doing have this, you know, type of setup of AI that can do all of this task and use all these tools, and—and you're more, you know, less of a—it's less of a hassle for, uh, anyone to kind of start using this because you don't know what to ask sometimes, and you just kind of have this question, and you're, you're thinking—you throw a simple question that you're not going to get like the information that you want, but this goes like step beyond that; it actually, like what you saw, it will just crawl to your site, check out your bio, check out your websites, and, and try to understand you as much as you can and trying to grab all the startup ideas that you have, and it says—it visited your personal website, checked out your social media profiles—it's got a little to-do checklist that it goes through, and it actually goes and does it, so that—which is kind of, you know, remarkable, um, on where we are with these kind of tools, um, and I, I think this will help, um, people who are, uh, non-technical, and I think that, you know, AI tool like Manis AI will help close that gap, um, I think a lot of people were just kind of, I don't know where to start. I'm sure they watched all your, um, podcasts and all other people's, you know, YouTube videos, um, but some people still seem very lost, uh, as is what I noticed, and they're always like asking questions like, how, what do I, where do I start? Like, what can I do? You know, how, how do I ask? How do I prompt? What template to use? It's, it's sometimes it's just bit too much for some—a lot of the, you know, non-technical folks. I think that this closes that gap.
Yeah, I mean, it's non-technical, and it's also like—this is onerous and a lot of work to go and do all this stuff, so if you can have Manis running, you know, during the day while you're working, or even if it works at night too, I mean, anytime, I guess, um, it just seems like it could help you even if—I mean, it's could help you come up with the ideas that you want to do and then put it out in the world, allegedly. Let's see if it actually could do it. Um, by the way, I just heard my doorbell ring, and I was like, my first reaction is like, is Manis—is Manis at my door? Is Manis just like coming, you know—Manis is that you? Yeah, they'll be pretty wild, like, now a robot shows up at your door, and, you know, hey, I want to ask you a couple of questions about your startup ideas. They'll be pretty crazy. I feel like this guy's like stuck here collecting Greg Eisenberg startup ideas from his podcast, YouTube channel, and website. Maybe it's just like a lot—that's a lot—let, let what happens, uh, I feel like stuck, but cuz typically when I see using browser, I will see like, you know, navigation in the browser. Yeah, it, it was doing that before, and now it looks like it's just stopped. So could we ask Manis, are you stuck? Yeah, let's ask that. Checking out your videos too. Maybe Manis is like, I don't want to give this, this, these guys ask them the startup ideas; I'm going to go and like build these myself. I don't want to—are you stuck? Are you stuck, bro? Wow, there we go. Okay, so it says, I'm not stuck. You point the finger to YouTube page. Yeah, okay, blaming YouTube. Okay, man, my dog ate my homework too. Uh, let me continue researching Greg Eisenberg startup ideas from multiple sources and compile a list of his most promising concepts. I'll focus on gathering specific startup ideas. Okay, let's check back on DocuSign, see where is, where—see where he's at. So he's, he's creating the app now. So another cool thing about this is that you don't have to have a like a project structure, uh, as you see on other tools like where, you know, people say go clone this, you know, template, you know, this, uh, boilerplate, uh, code, and you can use that to build your app, and have, um, Cursor and, you know, Webflow, and other tools to, to, you know, build that for you because, because their, their template is more optimized; it's already like skipped a lot of steps, so you don't have to like ask Cursor to do it for you, uh, and this one you don't have to do anything at all; you just kind of say, just go ahead and create me this, uh, app, and it'll just create its own project structure and everything. Now I know it's similar to what, you know, Cursor and Webflow does as well, um, but it's just another, easier step for, uh, people who wants to get started with this kind of tool.
How, how worried are you that this is going to Chinese servers, and how do you think—how should people think about that? How is it going to turn out? No, no, like I know that Manis is a Chinese company, right? Y, so there is some concerns with people around, you know, having a Chinese—like giving data to Chinese companies. Should people be worried about that? How, how, how, how do people think about the fact that this a Chinese company, not an American company? Yeah, I mean, I think that there is a valid concern that around that. I think, um, the, the one of the biggest things is that if you are a Chinese company, um, I think you're obligated to give access to the government. So because of that reason, and I know people are very reluctant to use this, and, and I think, you know, again, it's a valid concern, and I'm, I'm a huge, you know, security, uh, data safety, person—privacy—so I, I don't, I don't want people to go and like just start throwing any personal information and give any, any kind of stuff on these kind of tools, um, but at the same time, I want them to see where other, you know, competitions are at. You know, if we don't know our competitions, uh, then we, we're going to be behind, and I think that again, this is a lot—some people are calling this the second DeepSeek moment for China, and I mean—I mean, starting to see a lot of that lately, uh, and I think that we need to get ahead of ourselves in these kind of tools, and I, we, we have the technologies; we just didn't have someone to put the technologies all together and package it like what Manis has done, and I think it's good that they have done it, and now that we know, you know, what can and, you know, cannot be done, and I'm sure there will be many US companies that's now, um, you know, going out there and trying to, uh, you know, build something very similar to this tool, which is great, and I think that, you know, need—we need options; we need, you know, different alternatives—and I know there are already several open-source projects that popped up, uh, that builds around what Manis has done, so I think that's great, and I'm, I'm a huge open-source, uh, supporter, and I, I use open source for almost everything, so I think that, you know, this—there's, there's nothing wrong with, you know, using the tool, uh, that is located in China—just be very careful and mindful of, you know, what type of data that you give to and, and give access, access to, because some, some of these may ask you on certain things, and you don't, you don't have to give those informations—just keep it big—um, just, you know, do simple stuff, like try it out, like go clone this website or something and just kind of see, uh, how the tool works and what you can get out of it. I think that's the important thing—the learning, um, you know, I'm a lifelong learner, so I'm always trying out new tools and trying on new technologies, uh, so people should not be scared to try them out, but always be mindful of the data that they share. Cool, that's fair. Yeah, I mean, all right, let's, uh, let's see where we're at with, uh, the Greg Eisenberg startup ideas.
Yeah, he's still stuck trying to collect the data about you, so—and this happens, um, it's, it's not like perfect; it's in—it's still in beta, um, there are a lot of, uh, errors still, um, but it's still cool to see that how it was browsing your websites and all that stuff, which is neat, and you can go back, and you can't go back and see what it was doing and, you know, all the search that it got—it found your website, your, your X page, your YouTube channel, your LinkedIn, your Instagram—it's scraping all your social media profiles and checking you out basically because they want to understand you and want—me—because I asked—I specifically said your name, so it wants to know who you are and all your startup ideas, so it's going to go and like just try to find everything. Okay, wow. Manis computer has encountered a critical issue. You can choose to reset Manis's computer or start a new session, so what should we do? And this does happen, so, um, when it, when something like this happens, uh, you could either opt to reset computer or start a new session, but let's try to reset the computer. So it's going to try to reboot this sandbox and try to restart the task. The cool thing is they remembers where it left off, um, now it's not a guarantee that this, it will, will be able to, uh, fix whatever the issue that he ran into, so we'll see. So it's going to reset. So let's go back to DocuSign and see where it's at. Still doing his thing; it's still building his, uh, projects. Um, I like—it's, it's kind of, you know, very like worry-free; like it's doing all the stuff for you; uh, it's just—I just sit back and just kind of relax. Would it be annoying to ask Manis like, hey, Manis, what's your ETF? Yeah, let's try that, see what he says. Just like every annoying boss. Yeah, but remember the AI is a—what is the word—ever, uh, like not annoyed intelligent assistant, right? Yeah. Okay, so—oh, it says it's—I don't know if you're like—it was trying to troll us or something. As soon as I asked that, he was like, I've completed—it's like, I'm done. Like, what's your ETF? Like, oh, I'm done. Like, so were we like waiting this whole time and just until I ask—I'm like, okay. All right. Would you like me to deploy? All right, so now I ask you a questions like, do you want me to, uh, deploy this to the public network? I say yes, so—and what is that—what that's going to do is it's got its own public, like similar to what like Bell Replit has, uh, it's got his own little deployable website, and it's going to give you a subdomain with their, uh, primary domain attached to it, um, so you can actually see the, uh, website in action, and it doesn't make you sign up to any service when you do this, right? Does that work? Not yet. Yeah, not—I think that when you start asking questions like, and I, and I haven't gone to—
A lot of questions and so sort of like an iterative approach to getting to your results. So I think you know that’s that’s sort of the step I’m taking with this, and and that’s the sort of step I take usually. Um, for my audience as well, I I don’t try to overcomplicate it, um, and I think that’s sort of the sort of the way these agenty AIs are going towards. Um, I think it’s trying to take a lot of that load of thinking off you, um, and trying to understand, you know, what you want. Um, and maybe in the future, maybe they’ll just know exactly what you want after a while, you know, like after working with you for a long time, it will probably already know your history of what you wanted and you don’t even have to tell them again. I think that’s of like the AI future that we’re going to start to see.
So you got me thinking of something I think that could be big, and I don’t I don’t think we’ll have time to cover it today, maybe we will, maybe we won’t, but this Manus seems to me that it’s it’s it’s not just about how can you go and build a product and launch it, but it’s also probably very useful for growth tactics and actually marketing and growing whatever it is you built. So I’d imagine that you probably could do things like SEO. I’d imagine that you probably could do things like uh help, you know, create a content calendar, writing, researching based on different styles, you know, getting basically organic content, that sort of thing. Am I right or wrong on that, man?
Yeah, yeah, I I think absolutely you’re, you know, right on point on that one, and I think and I just got me thinking maybe we should, you know, have Manus go to one of your sites and just see, you know, how is my SEO on this website. Great, let’s uh let’s pull up latecheckout.agency because I know the SEO on that’s probably garbage; it’s literally a landing, you know, quick framer landing page. If you go to latecheckout.agency, it’s our design agency. Well, I guess you don’t even need to go to it; we can just ask Manus to go to it, right? Okay, yeah, this this website. Okay, all right, let’s ask him, let’s take this URL and say, “critic,” or I want to… Here’s how I would I would say: “Like I have a feeling that my design agency could be getting a lot more SEO traffic; I’m looking for help from you.” Oh, I’ll wait till you type. Okay, “Help me improve SEO for my…” Let’s start a new session. Yeah, see, I’m a long-winded prompted prompt guy. I’m a long-winded prompt guy, like I I write novels, you know what I mean? Oh, really? Yeah, yeah, I’m one of those. I try to keep it simple. Yeah, I like it. Okay, so you’re gonna say, “Help me improve… get… not not my website, it’s my design agency, this one.” Yeah, okay. Yeah, it’s yeah okay, that’s fine. Let’s remove your name, maybe. Yeah, exactly. “Let me improve SEO, but I want to be clear; I do want to be clear that, you know, basically this this firm, LCA, Late Checkout Agency, it it works with like the Dropboxes and, you know, the Grammarly and the, you know, Character AI, all these software companies and and and sort of big companies that are looking to be AI-first and transition to be AI-first and building AI products for them. I want to… it’s not just that I want better SEO; like I want the executives from these tech companies to come to the website, right?” So you want those tech tech companies come to this website. You see all the companies we work with: Character AI, Bolt, Salesforce, Slack, and look at all the executives we work with at those companies. I want more of those executives to come to this website via SEO. I think it’s important when you’re doing SEO, not just to do it… doesn’t… if I got a million visits today to this, I don’t really care; I’d rather a thousand visits, but they’re highly targeted. Yeah, there we go, you said it better than I could. So he, for the listeners on on Spotify and Apple, I want a more executive uh of AI-first companies to come to this website.
Okay, so it says, “I understand you want to attract more executives to Late Checkout Agency’s website. I’ll continue my SEO analysis with the specific goal in mind.” Very cool. I love how he’s like running into… like we’re we’re software testing Manus AI in real time. Totally. Again, they’re going to remind us that uh oh, it’s in beta. Yeah, yeah. Our bad, we’re just testing. I I believe a lot of the softwares are all in beta, and we’re just software testing for them, so there’s no news there. They just they just removed the label “beta,” but there there’s a lot, you know, I I work with many software companies, and uh there’s a lot of issues uh behind the scene that we just don’t talk about. So all right. Oh no, it has run into an issue. This guy is not going to deploy. Oh, so it’s so this this is a new uh message for me because I I I guess I haven’t gone to very long contexts, but Manus’ performance may decline with extremely long context. Um, so that’s what it’s saying. So I think it’s running into a context limitation. So that’s one of the things that I actually wanted to talk about on this episode was that uh there are some limitations uh to Manus; it’s not, you know, completely magical and just works like everything, and it’s going to do everything for you, and you don’t have to worry or anything. Um, the limitation I’ve run into obviously is the context. The context has uh like, especially if there’s like a becomes a very complex task uh and it and there’s a lot of information that it has to process, uh it does run into this issue, and I think that this is probably the biggest um issue that Manus AI team needs to probably address uh before it goes uh it goes live uh because I think that uh a lot of people are going to run into this similar issues, especially power users like me. I’m going to, you know, keep pushing it to its limits, and we’re going to run into this type of show all the time, and and this is not the first time I’ve seen the U the error that is is it’s gone into other than this context. Um, that it does it does ask you like, “Oh, I ran an issue; I need to reset the computer; I need to do this and that,” and you know, just ask me if I want to uh go ahead and restart or start a new session. So basically, starting a new session means like it may have run into like some kind of long context issue that it’s not able to process, so you have to start like a whole new session, and that’s kind of the down then the negative part about this tool is uh is is the context size.
So far, good to know. Yep. So we were not able to complete his stock design website. So that’s a bummer. Um, I think that we probably could do a local deploy and try. Yeah, let’s try it. I mean, even the fact that we can do a local deploy is absolute insanity. Actually, it was trying to zip this thing for me for me to download, and it failed earlier, so let me check let me check if uh it was able to finish uh creating the uh the zips. I mean, it’s created all the document structure, the code files, but it doesn’t have the the zip. Yeah, it’s so interesting to see how different people prompt. Like I I would be prompting right now; I’d probably say, “I’m just trying to get this to a local… to deploy this locally. I want to create something that’s beautiful that I can show on the podcast, um, and do you think that you can make that happen?” Like that’s how I… cuz I talk as if I would be talking to a coworker. Yeah, I think that, you know, um, it’s probably quicker anyway to run it locally to see the results right away because like deployment is is not as straightforward with these tools, um, and I know Replit has has done that really well, um, and I think I think that’s that’s a it’s got some good future in, you know, where the direction they it’s trying to go. Their AI agent is is is it’s getting better and better, um, but uh Manus has to work on their deployment process as well. So I think I think they’re just trying to do too much. I think that’s part of the problem is they’re trying to have all of this kind of package, like all this services and products are kind of packaged into this one thing, um, including deployment, and and obviously it kind of shows that deployment is not an easy easy one that they have solved yet, um, in my opinion. I I know there has been like other very simple ones that they they’ve done uh that that shows uh this like if I do like a game, like you know, make make an airplane game, like peterfly.peter.com, right? Yeah, fly.peter.com. You’re talking about the Peter Levels? Yeah, yeah, yeah. So Peter Levels is basically made like a flight simulator using AI, Three.js. Oh, maximum daily usage limit. Yeah, there’s a yeah, so there’s a limit; I believe it’s 10, so I I actually used a few before this episode, so we can’t do this one today, unfortunately. It’s… you know what’s also cool is like it feels like we have like four employees working at the same time. What’s that? It feels… it feel… what’s cool about Manus is it feel… it really does feel like you have multiple employees working at the same… it’s like, “Oh, we got this SEO, this marketing person; we’ve got this researcher; we’ve got this cracked PMD developer.” Yeah, yeah, yeah. You got one guy working on DocuSign, you know, you know, one agent working on your your startup ideas and one, you know, working on your uh on the, you know, improving the SEO for your LCA. So it’s crazy. But um, another thing is that the uh the problem that I see is scale. So obviously, you know, AI scale is a challenge for a lot of the companies, um, and the reason why they’re very uh right now they’re invitation-only uh and with an invitation code that you can only have access to Manus, um, and they’re they’re limiting those codes too, and I and I’ve been privileged to get like 10 invitation codes from them. Um, I actually given that out uh to to users to my um my X account. Um, I actually asked for 10 more; I was gonna I was going to offer it to your audience, say, “Hey, Greg’s audience, you get 10 codes if you want to try it out,” but didn’t happen. So um, I think that uh the thing that I noticed is that very early on when people were starting to get access to Manus, um, their we, you know, their their web uh has been hammered, um, and the reason why we’re seeing a lot of errors today, actually, I was not seeing this this much error uh in the beginning, um, because there weren’t as many people who had access, but now that that uh more access has been given out and more people are on the system, uh I start to notice a lot more frequent errors, and I think that’s part of the um scale issues that they’re going to have to address, um, again for before they go live. Um, and do we know when they’re going live? I don’t know yet; uh I have not heard any… Yeah, and I think that they, you know, they’re trying to tease early. Uh, looks like they launched premier uh kind of like what… it’s been like six days now. Yeah, they they they actually launched it like six days ago, but it didn’t like go viral until like two days ago, basically. So when it came over to us and people in the US start to get access to it, and that’s when it start to get like real traction, and I was one of the early ones who got access, and and it it was working pretty good, um, you know, obviously we’ve done I’ve done stuff like um pretty basic, you know, comparisons like doing the deep research or deep search type of thing, making a… and as you can see on the left side, you know, I I I I had more, but I deleted them because they they kind of crashed, but you know, build a simple SaaS website, you know, clone to apple.com. Obviously, apple.com is one of the, you know, well-designed websites, so um, I think it’s a it’s commonly asked thing, like people like go on apple.com, uh, and then, you know, do a casual multiplayer game, um, but the uh the research one works really well, obviously, um, you know, you expect that to work well, um, simple games work well, uh landing pages work well, um, and I think that’s sort of the ones that have been very successful uh with Manus AI so far. Um, I think that the other one that it does well um is something like this, like you’re uh analyzing your websites uh for optimizations, um, scraping data, uh it can actually because is able to control the browser and go to go to the website and actually crawl through it like clicking stuff, like you can actually like scrape uh data uh using it. So I think that’s one of the uh very easy easy to use and, you know, probably frequently used uh tool to kind of get information for uh for the people. And the as you get more complex uh like DocuSign and stuff like it does it does have its challenges, but I I like to, you know, push things to its limits, so uh just to wrap up, maybe let’s just go through what we’ve done today. Let’s go through the DocuSign; let’s go through the startup ideas, and let’s see let’s see what we were able to do in, you know, 60 minutes and using Manus.
Yep. So we started with this, “Create DocuSign clone,” um, and middle of it we say, “Hold on, let’s just focus on the e-signature,” so we we say, “Okay, just focus on that,” and it replanned his um project, and it set up all the code, and I didn’t actually get to see the code yet, but it looked like it set up its typical Next.js uh app, set up its routes, um, and and then it was create an interface with login page, register page, the routes, the passwords, the middleware routes, um, so it came… it looked like it went through a lot of the um important things that it needs, the basic ones at least, um, and once it got through creating this uh code, uh it got stuck on getting the deployment uh successful. So we got document upload, document management, and the signature field placement. So I’m actually excited to see if it actually runs, but it wasn’t able to deploy, so I what I would do is um I would I would actually like run this locally and then I can like send it over to you, see what that looks like. Um, again, uh one thing that I I do uh when it asked me to run it locally is uh I do have to scan through the code, just just in case. Totally. Yeah, so I don’t just download this; that’s why I didn’t want to do it on the uh on this recording, um, because that’s fair. I didn’t want to look like paranoid; keeping it real here, huh? Trying to play safe, you know, I don’t know what’s gonna happen, and I do like to create this into like a my own sandbox. I don’t I don’t download this stuff into my local computer; I have a I just create like a virtual sandbox on in the cloud, and then I I run it there, just in case as well. So I don’t want to do that on this uh session, um, but I I will do that, and then I’ll show how that ran locally, um, and then let’s go back and check your um yeah, top startup ideas about highest potential. Again, it got stuck. So the thing with this one is interesting is because research has been very straightforward for me, and it worked, uh, and like I was uh mentioning earlier that since people got access to this application, um, it’s I’ve seen a lot of these issues where it it hits a wall, um, and this should have been a very easy um easy thing that should have completed, um, but it wasn’t able to. So and again, like probably the time of the day is not the best; I think a lot of people are probably on it and asking stuff to do, so you may be running into more of this issues. Uh, I may try to run this again later, like late late in the night, um, whereas the traffic a little bit lower, um, and and then the other one is that we try to ask to create an airplane game, like peterfly.peter.com, in Three.js, deploy in public, and and you we reach the uh maximum daily usage limit. Um, again, this is in a beta stage; there’s a I believe there’s a 10 request or 10 sessions um limit in the day. Um, the cool thing is that even though it says this, I can actually go into um another session and run it; it’s I believe it’s the sessions that it matters, so I’ll I’ll use the one that is like broken, which is like this one, I think, right? This one’s this one’s still working, so this one’s broken, so I I’ll use this one to say… I don’t know if this is a loophole that I found, but uh when you reach a maximum limit, just start a different session, and it will… you can’t tell it to do something else.
That’s awesome. Yeah, so now see, it’s like, “I’ll help you create an airplane game,” so it’s just like a new session basically, but the only only difference is that it’s got like all this other stuff in the context, so hopefully that doesn’t interfere with what I just asked, right? Oh, might Victor from Hugging Face uh actually did something similar, but he he didn’t specify fly.peter.com, but he just say create a, you know, flying airplane uh game where I can control the airplane and whatnot, and it just kind of… it looked like he did it in one shot, um, and I know people are asking like, “Was this one shot, or was there like follow-ups?” But um, I’m sure there was a followup; it it it doesn’t actually like produce everything like in one go sometimes, and you may have to kind of tweak some of the uh the features, uh especially like um aesthetics, I noticed. So you may just like throw like a very simple aesthetics, and you may have to say, “Oh, can you make this a little bit more uh pretty, or make it more… I don’t know… more objects in the scene,” uh cuz if he just asks you to like create like a flying game, it may just create like a sky and a and like a plane field, and you you may have to give like specific instructions on: “Add a building, add a tree, right? Add lakes or something, something to make it more interesting.”
Okay, so yeah, we might we might even show a clip of of that one-shot prompt, so just showing what that could look like. So Manus sent me a message, literally this is just right after the podcast was recorded, saying our uh flight game had been created. So we basically one-prompted a flight game; it looks like a more basic version of Peter Levels’ game, but it’s pretty cool. Like here you can see the plane flying, and I believe you can uh shoot… well, maybe you can shoot, but that’s the idea of the game, and you shoot down these balloons, um, which is which is pretty cool, um, and you can play it; it’s deployed; you can play it on your web browser. I think the next step for me would be… how do you… I mean, just make it look a little more visually cool, um, and also add multiplayer. I think one of the reasons why Peter Levels’ uh game… I mean, has gotten… there’s a few reasons why Peter Levels has done 67,000 of revenue in the last 30 days; part of it is on his his audience, you know, he he has a big audience, so he can distribute it, but the other part is it it is fun. I know I tweeted about his game, and a lot of people were like, “It’s not… it’s such a stupid game,” and it’s like, “I don’t know; I was playing it for 30 minutes, and I was enjoying it. Like I could have I could have watched The White Lotus, but I was watching I was playing that; I enjoyed it.” So um, Manus did it; it took one one little prompt, and we got a game working, uh, so that’s really cool, and it was deployed. So we we’ve gone over the DocuSign; we’ve we’ve gone over the game; we’ve got the SEO, um, and we’ve got the research assistant. We’ve done this all in like an hour, which is kind of crazy. Yeah, we haven’t been able to successfully put deploy, you know, we we basically deployed the app basically, but it it got us almost there; like we just have to review the code, deploy it, or, you know, run it locally, make some edits, and then finally deploy it. But um, overall my take on this Manus, honestly, is that it’s yeah, it is the closest thing I’ve seen to AGI yet, and it seems it seems like it’s close; it’s not there yet; like I need it to I need it to deploy, but the fact that it feels like an operator mixed with like a, you know, Replit mixed with a ChatGPT, like I think that having having this is is super super super helpful for people who want to like automate, build their businesses, which, you know, and that’s why I had you on the podcast. So Manus, I want to thank you for for coming on the podcast, and I want to ask you, you know, where could people find you? We’ll include those links in the show notes.
Yeah, hey, thanks for having me, and this was uh fun, and people can find me on x.com/minoy. U, that’s currently my main um page. Cool. All right, we’ll include that. Manus, thanks for… I mean, you had the beta access; I didn’t, so thank you for for for hooking it up for everyone. I think a lot of people are appreciative of that, and uh let’s go build some businesses, and I got a I got an extra key for you so that you can have access to if you want. Oh my God, yes, please do, and I’ll play around with it. If people want more Manus content, please do let me know, um, and uh I’ll see you next time, Manus. All right. Also, Manus, I know this is your first podcast that you’ve ever done, so thank you for sharing it with us. Hey, thanks for thanks for inviting me; uh I’ve always meaning to do one, um, just didn’t get to that step yet, uh but maybe this will be uh sort of uh my kickstart into the podcast world. Absolutely. All right, catch you later. All right, man, baby.