AI transcript
0:00:05 Hey, welcome to the Next Wave Podcast. I’m Matt Wolfe. I’m here with Nathan Lands. And
0:00:11 right now, agents are all of the rage. We are in the era of AI agents. And right now,
0:00:15 the biggest agent that everybody is talking about is Manus. So in this episode, we’re
0:00:19 going to break down what is Manus. We’re going to show you what other people have created
0:00:24 with Manus. We’re going to use Manus ourselves and actually build a really cool website with
0:00:28 it. You’ll see what it built before the end of this episode. We’re going to break down
0:00:33 how Manus was made, the bigger implications, how you can use it in your business. This
0:00:38 episode is jam packed with AI agent goodness and how it’s going to impact your life and
0:00:42 your business. So get ready. You’re going to be in for a fun ride. Let’s get right into
0:00:50 it. HubSpot just dropped their 2025 marketing trends report, and you’re going to want to
0:00:57 see what they found. Visual content is delivering 21% more ROI. Small influencers are building
0:01:04 45% more trust and AI is changing how fast we can create. And the best part, it’s not
0:01:10 just another report full of stats. It’s a game plan you can actually use. There are frameworks,
0:01:15 AI guides to help you do more with less and real case studies and playbooks from teams crushing
0:01:22 it right now. Want to nail marketing in 2025? Go to click HubSpot.com slash marketing to download
0:01:32 download it for free. So where should we start? Maybe we should just start showing off some of
0:01:36 the stuff that other people have made with Manus. Yeah, I think there’s been so many crazy posts on
0:01:40 X like showing how people are using it. I think let’s just start there and show what people are
0:01:46 doing with it. You and I have both played with it a little bit already, right? Like I shared a screenshot
0:01:53 on my X account of me trying to play with it. And in this tweet, I basically gave it the prompt to go
0:01:59 research Manus because I want to make a video about Manus. So go research Manus. What is it good at?
0:02:05 What is his limitations? What’s the competition? You know, just give me as much deep dive info as you
0:02:13 can on Manus. And it broke right during the portion where it was investigating the limitations of Manus.
0:02:20 So literally the spot that it decided to stop working was when it was starting to deep dive on
0:02:25 its own limitations. Yeah, that makes sense. As intended. But anyway, I know you’ve tried to play
0:02:30 with it as well and sort of ran into some… No, I just tried to play with it a little bit. Like it took
0:02:35 forever to get access. You know, I had to message a bunch of people on X to get access. Then I finally got
0:02:39 access and I don’t know if it was just the time zone difference in Japan. Maybe they were doing
0:02:44 some maintenance or something. But when I tried it, like it wasn’t working. So I haven’t actually got
0:02:48 to use it yet. I tried it one time, didn’t work. And so I was like, well, you know. Yeah, yeah, yeah.
0:02:53 Well, I think what’s happening is that Manus is insanely overloaded. They actually tweeted this
0:03:01 on the day that we’re recording. Seven days, two million people on the wait list. So yeah, I think,
0:03:05 you know, they’ve been letting a lot of people in every day. I think they’re letting like,
0:03:09 I don’t know, 50 to 100 people in a day or something like that. And I think they’re already
0:03:13 overloaded. So I think that’s what we’re seeing right now is just so many people trying to use
0:03:19 it at once. Even that one that I was just showing you here where it says investigate and document
0:03:24 Manus AI technologies limitations. And that’s where it actually stopped working. If we look over here on
0:03:32 the left, it says high system load has caused an internal server error. So it didn’t like get messed up
0:03:36 because it didn’t want to, you know, share its own limitations. It got messed up because
0:03:42 at that point in its own processing was when the servers got overloaded and it just sort of stopped
0:03:48 working. But yeah, let’s take a look at what some other people have made here. So this one is from AK
0:03:57 and he actually got it to make this like instant run game, very similar style to that, like levels IO
0:04:03 flight game, a lot more basic, obviously. Yeah. But if you’re listening on audio, it’s basically a
0:04:09 character just like running in a straight line and it’s like a 3d world, but everything’s sort of like
0:04:14 basic shapes. It’s like squares and cones and stuff. It’s very, very basic looking, but it’s just a
0:04:18 character running and obstacles getting in its way. And it’s obviously trying to dodge the
0:04:23 obstacles and like grab power ups and stuff. But that’s just from a single prompt, right? So like
0:04:30 it went off and did everything. Yeah. Literally AK says the prompt was make a three JS endless runner
0:04:36 game. Yeah. So one, two, three, four, five, six word prompt made that. Yeah. And he specified three
0:04:41 GS and it’s kind of interesting. Like I’ve played with three GS for like years and it just never took off
0:04:45 really. And then it’s crazy now that it’s taking off because of AI. Well, give me some more context.
0:04:50 What is three JS? I’m a total non-coder. I think it’s like a JavaScript library for like 3d stuff.
0:04:55 Oh, okay. So here’s one from our buddy, uh, Bilal here. He says I tested Manus AI. It’s the coolest
0:05:01 thing I’ve experienced to a truly autonomous AI agent. I can’t wait till this thing can use desktop acts
0:05:08 like premiere and Photoshop. It low key feels like baby AGI. That was actually my first thought. Like when I
0:05:14 actually was using Manus the first time, it felt very similar to baby AGI. We’ll show it off in a
0:05:19 minute if it works, hopefully. And what it’ll do is it actually builds out a plan for you. And then
0:05:24 once it builds out the plan, it starts back from the top and then tries to execute each step in the
0:05:30 plan, which feels very similar to what baby AGI was doing in the past. I mean, baby AGI was a great
0:05:34 concept when Yohei created it, right? It was just, I think the models were not good enough to actually
0:05:38 execute on those plans. It’s exciting to see that now they’re finally getting good enough to actually
0:05:44 start executing on, on the plans they generate. Yeah. Yeah. For sure. Yeah. I mean, when baby AGI
0:05:48 came out, it would get stuck in these loops. Yeah. It would like go to step three and then for some
0:05:52 reason, just like keep on searching the web and never get past step three. Yeah. This one actually
0:05:58 seems to, you know, actually follow through and finish, but here’s the prompt he gave. Can you open
0:06:04 up a 3d view inside Google maps of Austin, Texas? And let’s find the most compelling photos to take
0:06:10 of the downtown skyline. It’s very important that you change the map perspective so we can get oblique
0:06:18 3d views versus a top down one. I want you to fly around surveying downtown from 3d vantage points
0:06:24 in Google maps and tell me your favorite location slash photos. So that’s the prompt, which is,
0:06:30 I mean, Bo Aval uses big words. I love him, but he does, but you can see here very similar to baby
0:06:37 AGI. It went and created this task list for the agent to follow, but you can see it says research drone
0:06:41 regulations in Austin, search for potential drone locations near downtown Austin, create a short list
0:06:48 of locations, scout locations using Google maps, 3d view, evaluate locations for drone flight suitability,
0:06:54 complete top three recommendations, prepare final report with recommendations, deliver results,
0:06:59 to user. So that’s the sort of task list that it gave. And then you can see as he scrubs through,
0:07:05 it goes through, it does all these tasks. We can see here, it’s actually looking on Google maps for him,
0:07:12 trying to find locations, and then eventually gives him top three recommendations, why it’s recommended.
0:07:17 So, I mean, that’s pretty wild that it can even do that in my opinion.
0:07:24 And then we’ve got this here. This is a thread from Elsine over on X. And this is a thread of a whole bunch
0:07:31 of different shares that he found. So the first one, he had it clone the Apple website. This one seemed like
0:07:37 a simple task to me, but we can see here, it literally made like a exact duplicate of the Apple
0:07:38 website here.
0:07:42 Wait, did it take a screenshot of the Apple website? Is that how it was generating the website?
0:07:42 I’m pretty sure.
0:07:46 It looks like it actually like took screenshots of some of the elements, right? Just like pasted them
0:07:47 in, which is hilarious.
0:07:52 But then we’ve got an analysis of Tesla stock where it went and did a whole bunch of research on Tesla.
0:07:58 I don’t really feel like you need Manus for this. I actually did this exact one. I had it go and
0:08:04 analyze Tesla stock for me using ChatGPT’s deep research. And it was really, really, really good.
0:08:09 So I feel like Manus is almost overkill just to do this research here.
0:08:14 Yeah, you can use that, Grok, Perplexity. They’re all great for use cases like that already.
0:08:22 Yeah. Plan a family trip again. Do you really need like an AI agent to plan the family trip? I mean,
0:08:27 an agent like Deep Research, you know, from ChatGPT or from Google or from Perplexity,
0:08:32 like those deep research agents are really good at planning family trips. I don’t really feel like
0:08:37 you need Manus to do that. The ones that I think have impressed me more have been more these like
0:08:43 visual ones. Like here’s this game that somebody created again with 3JS. Yeah, we can see the tasks
0:08:48 that they did to make this game create project structure for 3JS playing games, set up 3JS
0:08:53 rendering environment, design and create plane 3D model, simplify left, right plane controls,
0:08:57 fix plane shooting, test, improve plane movement and controls, finalize game implementation,
0:09:01 share updated game version with user. And then on the right side of the screen,
0:09:06 we can actually see the plane game kind of look similar to, you know, Peter Level’s game and that
0:09:13 game we saw from AK similar style with where it’s real basic shapes, but it did it with a single
0:09:20 prompt. Website SEO optimization. Again, not really a great use case for Manus in my opinion.
0:09:26 Double agent productivity. Somebody actually showing they were driving their Tesla and they had Manus
0:09:32 running on the screen in their Tesla doing some autonomous stuff. This one was actually shared by
0:09:36 Manus. So it’s a little bit more legit. Double agent productivity while Tesla’s full self-driving
0:09:42 takes a power user to his coffee meeting. Manus preps him with all the key talking points. So he’s on
0:09:48 his way to a meeting while Manus is rounding up talking points for him to actually talk about at the
0:09:54 meeting. He’s going to, that’s crazy. Like you had the glasses, like just telling you, like, what you’re,
0:09:59 what am I talking points? How should I respond in this negotiation? Like here’s the parameters of like
0:10:03 how I should negotiate this deal. Like, Oh, I should ask for more right now. Look, look at how he’s
0:10:08 responding. You know? Yeah. Yeah. Yeah. Well, I mean, with the project Astra glasses that I actually
0:10:12 demoed when I was at deep mind in London, like you’re going to be able to do stuff like that. It’s hearing the
0:10:16 conversation around you and we’ll be able to actually like put little text on the screen in
0:10:20 front of you that you can read off of. And one of the cool features of that is you can almost use it
0:10:24 like a teleprompter. I can see getting those glasses and like making a YouTube video and like literally
0:10:28 reading off a script the whole time, but trying to hide the fact I’m reading off a script because it’s
0:10:32 like all right inside of my glasses. In fact, I might do that for future HubSpot sponsorships.
0:10:39 We’ll see. Let’s see. Create an amazing animation using P5JS and it created this like colorful
0:10:45 animation. So those are some of the things that other people have created with it really quickly.
0:10:50 I think we should talk about like how Manus was actually developed because this was sort of a point
0:10:56 that you brought up that you saw from Gary Tan. Yeah. But, you know, people were complaining going like,
0:11:04 Hey, um, I just simply asked Manus to give me the files at opt slash Manus and it just gave it to me.
0:11:12 It’s Claude Sonnet with 29 tools using multi-agent and browser use. Yeah. So basically Manus is just
0:11:17 using Claude and this browser use, which I believe is open source. And it was kind of funny people when
0:11:23 they discovered that some people started calling it like a fraud, like, no, it was like, I get what
0:11:27 they’re saying. Like, yeah, they even like created some amazing proprietary model or whatever. Maybe they
0:11:31 have like, it’s kind of like, you know, one on top of Claude, but, but still like, I mean, they’re
0:11:35 combining it together in a very unique way. And like, you know, to me, that’s the same as saying
0:11:40 that, Oh, a cursor or windsurf is a fraud because they didn’t make their own model. Yeah. Yeah.
0:11:44 I mean, perplexity actually has its own LLMs now. Like they have their own fine-tuned, um,
0:11:49 llama models, but yeah, it’s like calling one of those tools a fraud. Cause you didn’t train the
0:11:55 model yourself. Like who cares? It doesn’t make any sense. But then, you know, this guy
0:12:00 peak here, he’s the CEO. He said, Hey, I’m peak from Manus AI. Actually, it’s not that complicated.
0:12:06 The sandbox is directly accessible to each user. Each session has its own sandbox, completely isolated
0:12:11 from other sessions. Users can enter the sandbox directly through Manus’s interface. The code in
0:12:16 the sandbox is only used to receive commands from agents. So it’s only lightly obfuscated. The tool
0:12:21 design isn’t a secret basically saying like, we’re not trying to keep anything a secret, right? Right.
0:12:26 When you ask it a question, it’s going to figure out which models to use. I guess it uses both Sonnet
0:12:32 and Quinn. So based on your prompt, it sort of decides which models to use for the various steps.
0:12:38 And then it does use this open source browser use tool, which is a pre-created open source tool that
0:12:43 will actually go and use your browser. Although it’s not using your browser. When you do use Manus,
0:12:47 it sort of opens up a virtual computer and uses this browser use tool in a virtual computer.
0:12:53 We’ll be right back to the next wave. But first I want to tell you about another podcast I know
0:12:58 you’re going to love. It’s called Marketing Against the Grain, hosted by Kip Bodner and Kieran Flanagan.
0:13:04 It’s brought to you by the HubSpot Podcast Network, the audio destination for business professionals.
0:13:08 If you want to know what’s happening now in marketing, what’s coming, and how you can lead the
0:13:12 way, this is the podcast you want to check out. They recently did a great episode where they show you
0:13:16 how you can integrate AI into the workplace. Listen to Marketing Against the Grain wherever
0:13:18 you get your podcasts.
0:13:26 Yeah, I think it is interesting that this year, like all the top American AI companies have been
0:13:32 trying to push out solutions with agents and they haven’t really got much attention yet. And then
0:13:35 this Chinese, which is not open source. Yeah, they said it’s gonna be open source, but this Chinese
0:13:40 company comes out and just combines all these tools in an interesting way and they make it happen.
0:13:45 So I think that’s gonna be a dynamic we’re going to continue to see. I should show the Gary Tan tweet.
0:13:50 Yeah, yeah. So here’s what the tweet says. It says, if real, this actually emphasizes the models
0:13:56 are plenty smart already and all the alpha is in custom prompting, tool use, and clever workflow
0:14:01 and evals. Yeah, I think that is probably where things are at. Even though OpenAI has got a great
0:14:05 model and Claude and everyone else, man has figured out how to combine these together in an interesting
0:14:10 way first, you know? Yeah. Which is exciting because it feels like all these models, they’re just waiting
0:14:13 for people to like build things on top of them, like new things and try new things.
0:14:19 Yeah. Yeah. Again, I don’t think it’s a bad thing that they went and used existing large language
0:14:24 models and existing open source tools. They’ve just combined them in a way that other tools haven’t
0:14:29 yet because they went and combined them in a way that others haven’t. They made something really
0:14:36 valuable and useful to people. And like Gary says here, it just shows these models are a lot better than
0:14:40 people are giving them credit for. Right. Right. It shows that these models are already capable of
0:14:45 going and just handling work for you. You just need to sort of have the right workflow and the right
0:14:50 combination of large language model and tool use to get done what you want to get done.
0:14:56 Yeah. I do wonder like, why did Claude not do this first? I wonder if it’s out of restraint kind
0:15:00 of thing, or they just didn’t come up with it. You know, I’m not sure which one it is.
0:15:04 They had, what was it called? Claude computer use, right? Yeah. Or something like that.
0:15:09 Claude was kind of the first one to push something like that out. Yeah. But it was such a pain in the
0:15:14 butt to actually use. Right. You had to install Docker on your computer and then you had to run a whole
0:15:19 bunch of like terminal commands and install a bunch of stuff. And like, even if you’re not that technical,
0:15:24 it’s pretty easy to follow the directions. It was just a lot of steps that I think scared a lot of
0:15:28 people off. Right. So nobody ever really used Claude’s computer use because it was like
0:15:34 too many steps. Yeah. And then ChatGPT came out with their operator, which, you know, simplified
0:15:38 things a little bit. Oh yeah, that’s true. And it’s actually doing something similar to this. It’s got
0:15:42 like a sandbox where like, instead of you having to put it on your own computer, they’ve got like a
0:15:48 virtual machine running. Yeah. But I guess, you know, Claude 3.7 is just, you know, that was,
0:15:54 that was the magic, right? That was the one that made it work better. You know, operator launched a few
0:15:58 weeks ago, maybe a month ago. I don’t remember exactly when, but it’s been out for, you know,
0:16:03 three, four or five weeks somewhere in that range. And it was all hyped up the week that it came out.
0:16:06 And then I feel like nobody’s been talking about it ever since. Right. It’s just sort of like died
0:16:12 off. So it must not have impressed people that much. And now we have Manus and like, everybody can’t
0:16:17 shut up about it now. And it’s like, well, it kind of is similar to what we already have, but it just
0:16:24 seems to work better. It just seems like this combination of Claude 3.7 plus browser use, plus,
0:16:31 you know, occasionally using Quinn as well was like that right mix to make it feel valuable and make it
0:16:36 feel like it’s really working for people. Yeah. I mean, it does feel like open AI, you know, they’re
0:16:40 trying to do so many different things. Yeah. Like early on, they were doing AIR too, which apparently
0:16:44 they have a big update coming or whatever, but you know, mid journey and Leonardo and other people are
0:16:48 like totally dominated that space. And so it feels like maybe it’s the same kind of thing where,
0:16:53 you know, the core model companies, they’re trying to do so many different things. And so a company can focus on
0:16:57 one thing like, okay, let’s make sure that the, uh, the AI models know how to use a computer really
0:17:02 well or use tools very well. Uh, they can still win by just having like really good focus.
0:17:08 Yeah, no, absolutely. Well, let’s go ahead and try out Manus. Yeah. Hopefully it works and it’s not
0:17:12 overloaded. And then after that, I have like a few alternatives that I’ve come across. Like we
0:17:17 mentioned, there’s 2 million people on the waiting list. So I thought it would be helpful to share some
0:17:22 other options that you can go and try out if you can’t get into Manus, but what do you want to test
0:17:29 here? Um, I mean, the easiest one would be like a website or a game. I think as of right now, yeah,
0:17:33 yeah. A website would be cool. I mean, that’d be good for like a business audience. I think a landing
0:17:38 marketing page would be probably, you know, a value for business people because you’re always making
0:17:42 landing pages and then pushing ads to push people to those landing pages to test them.
0:17:50 agents make that dramatically easier to do. Yeah. Create a marketing landing page for a website that
0:18:01 sells. What should we sell? Shoes? I don’t know. It sells shoes include social proof research on the
0:18:07 best shoes. I want to give it some extra details so it’s not just, you know, going off and cloning a
0:18:18 website. Yeah. Let’s see what else, uh, images of shoes and customer testimonials with headshots.
0:18:25 Let’s go use stock photos for now. Let’s see what it does. So it’s thinking if you already had images,
0:18:28 could you have provided that to it? Uh, there is an upload button. You can see there’s like a little
0:18:33 attach button down here so you can’t attach other stuff to it. Yeah. Let’s see. Uh, I’ll help you create
0:18:37 a marketing landing page for a shoe website. I’ll include social proof research on the best shoes,
0:18:42 images, and customer testimonials with headshots using stock photos. Let me get started on this
0:18:48 right away. So now it’s setting up a project structure, creating a to-do list, updating the
0:18:52 plan. And then I believe if I open this, yeah, we can see the plan that it mapped out for us.
0:18:57 So here’s the plan that it mapped out. Set up project structure, gather stock photos,
0:19:03 create customer testimonials, design HTML structure, add CSS, test and validate landing page functionality,
0:19:10 deploy landing page and share with user. So that’s what it’s doing. And then I can actually open this
0:19:15 up and we can actually see it taking actions in our browser right now. So what you’re seeing on the
0:19:20 right side of my screen is actually it just going hands-off. I’m not touching anything. That’s just
0:19:24 literally a video. This is the future of work. Yeah. This is literally just a video.
0:19:29 It’s watching the AI. It’s on the Vogue website. We can see every once in a while,
0:19:34 it sort of flashes bounding boxes on the screen. I think what it’s doing is sort of reading the CSS
0:19:38 and then, you know, pulling text from within certain divs or something, but it’s reading
0:19:45 Vogue.com right now, I guess about the best shoes. Okay. Hopefully it’s not just copying their website.
0:19:50 Right now it says it’s doing research. According to the task list, it says research best shoes and
0:19:55 current market trends for landing page content. So according to Manus, it’s just doing research.
0:20:01 It’s not, it doesn’t seem to be like screenshotting. It seems to be like using bounding boxes to like
0:20:07 find where the text is and then reading the text inside the box. Now it’s on a site called Birdie.
0:20:11 I mean, I assume this is not going to be perfect, but like you can see where this stuff is going,
0:20:17 you know? And even recently, I’ve already found that this has changed how I like hire contractors and
0:20:22 stuff. Like my own website, Lord.com, like I built it myself, but I was busy and I needed small tweaks.
0:20:27 I might occasionally hire that person to make tweaks to the website and stuff like that. And now I found
0:20:33 it doesn’t even make sense to hire that person anymore because like I just use Claude and just type
0:20:38 something in and it changed my entire website for me. Literally like in like five minutes, it made dramatic
0:20:42 changes to my website. I added like the next wave podcast, all that. I literally just told Claude to do all
0:20:47 that. Yeah. Yeah. Yeah. And it did all of it like five minutes and I pushed it live and didn’t have
0:20:52 to hire anyone. Yeah. Well, I mean, as you know, like I’m actually rebuilding future tools and yeah,
0:20:58 I hired a developer to help me build it. The developer kind of got really, really flaky and then started
0:21:02 telling me stuff that was originally planned was outside of scope. And it was this whole pain in the
0:21:08 butt process. I eventually ended up just like letting him go. I was out like basically 10 grand on this
0:21:12 redesign, but I went, this is just getting too insane. Yeah. It’s getting more and more expensive.
0:21:19 I’m going to cut my losses. And yeah, I mean, I’ve basically managed to completely rebuild future
0:21:25 tools with a better design, like just totally by myself using these AI tools. I think you’re going
0:21:29 to see this in like so many different areas of business. So, you know, I’ve been like doing more
0:21:35 and more game design stuff lately and been watching videos of Jonathan Blow, pretty famous and notorious
0:21:38 game designer. And he was talking about the difference between like a gigantic
0:21:44 company making a game or a small team or an individual making a game, how dramatically different
0:21:49 those things are. And he was talking about how so much is lost between like a designer or a business
0:21:53 person or an engineer all wanting different things or having different things in their head
0:21:58 and communicating that to each other is almost impossible. Usually there’s usually a lot of things
0:22:02 that are beyond just words and it’s really hard to express those. And maybe even impossible because
0:22:05 people have different understandings of different words and what you mean.
0:22:06 Yeah. Yeah.
0:22:09 And it ends up, you know, in a different result. I think you’re going to see more and more where
0:22:13 you’re like, okay, your idea of future tools, instead of you having to work with a designer
0:22:17 and working with an engineer and work with all these different people, you know what you want.
0:22:17 Yeah. Yeah.
0:22:22 And also you’ll know very quickly if you like what you’ve produced and if you want to change
0:22:26 it versus having to talk to all these other people who will never actually know what’s
0:22:27 inside your head.
0:22:27 Yeah.
0:22:30 What you want. And I think you’re going to see this in so many different areas where people are
0:22:34 going to be able to create amazing new things that, you know, whole committees of people
0:22:39 can’t really create, but you and your AI partner will be able to.
0:22:44 Somebody was asking me on X, like, but don’t you get like really frustrated when you’re trying
0:22:47 to code something and you run into a bug and you can’t get past it and it takes you forever
0:22:52 to get past that bug. And I went, yeah, I mean, that’s definitely frustrating, but it’s
0:22:57 not nearly as frustrating as having something that I don’t like on my website, going to my designer
0:23:00 or my coder asking him to fix it and not hearing back for seven days.
0:23:01 Yeah.
0:23:06 Because if I run into a bug while I’m coding and I’m struggling to get it fixed, it’s usually
0:23:11 like between 20 minutes to maybe like two hours of me going back and forth until it’s fixed.
0:23:15 If I was working with a developer, it could be two freaking weeks before I hear back from
0:23:15 him.
0:23:19 Plus an invoice because it’s out of scope, you know, the developers too, it’s like dealing
0:23:23 with a mechanic. Like if you’re like a hardcore developer, you have no idea what they’re saying
0:23:26 and if it’s true or not. And that’s actually another thing Jonathan Blow was talking about.
0:23:29 He’s like, he’s been that same dynamic where like a designer used to tell him something to
0:23:33 code and then he like disagreed with it, but he would pretend he agreed with it. He’d build
0:23:37 it in a different way. Then he’d make up an excuse why he did that, had to do it, you know?
0:23:41 And there’s all these disconnects where if you were able to use AI and be in total control
0:23:45 of a product, you will get closer to your actual vision for a product than dealing with
0:23:46 a whole group of people.
0:23:51 A hundred percent. I think the next like Stardew Valley or the next Minecraft is probably going
0:23:55 to come out of a single person just who’s passionate about making the game that they
0:23:58 envision and they’re just at home vibe coding it up.
0:23:59 Yeah, yeah, yeah, yeah.
0:24:05 But anyway, when it comes to the whole future tools website, I’ll just see this right now
0:24:09 in next week’s episode, I’ll break down my entire process to recode future tools.
0:24:12 So if you’re not subscribed, make sure you subscribe. I’m just saying that right now
0:24:17 while we’re waiting for our agent to go and finish up our stuff for us.
0:24:19 Waiting for it to do our work.
0:24:25 Yeah. So right now we’re on the third step out of about eight steps. It did its research
0:24:29 and now it’s gathering stock photos of shoes and potential models.
0:24:34 So it’s slow, but you know, these things will get faster and faster. Like probably they’ll
0:24:36 get like three to five times faster every year forever.
0:24:40 Yeah. Which means in like three years, we’ll have things like this where it’ll do like thorough
0:24:47 research and, you know, competitive analysis and check out your competitors designs and figure
0:24:50 out all that for you and put together and create a website in like a minute or something.
0:24:54 Yeah. Yeah. Once I see how it outputs it, I actually don’t know what the output’s going
0:24:59 to be because that’ll determine, you know, how you push it live. You know, I don’t know
0:25:03 if it’s going to give me like a zip file with all the files that created, or if it’s going
0:25:07 to actually like host it for me. Like I have no idea what the final output’s going to be.
0:25:12 I just know it’s not the fastest process in the world. I think once we get to the point where it’s
0:25:17 designing the HTML and CSS, it’ll speed up. I think the research is the slow part. I was just
0:25:20 looking at the terminal and it was actually like downloading images. I don’t know where it was
0:25:24 saving them to, I guess like the virtual computer. So it’s just copying somebody’s shoes.
0:25:29 Yeah. Well, I told it to use stock photos. So it just downloaded a bunch of stock photos.
0:25:33 It’ll probably have watermarks on them or something. Yeah. I can see now it just generated a bunch of
0:25:38 fake testimonials from Sarah Johnson, Michael Chen, Jessica Rodriguez, and David Wilson.
0:25:44 But now we’re on step four of eight. I do think once it gets to the HTML and CSS, that should speed up
0:25:49 because I know Claude is pretty fast at that. And if it’s using Claude, then it should pick up the
0:25:52 pace. So I guess we need to figure out what’s the key business takeaway for people.
0:25:58 In my mind, the key business takeaway for people is you could go and build the website,
0:26:02 that landing page, you know, that mini SaaS app that you’ve been wanting to build and you don’t
0:26:07 need to go hire a developer anymore. Right. Right. You can go and have a tool like Manus go and do the
0:26:13 research for you and build the bones for you. Like I’ve had software ideas over the years. I’ve been like
0:26:19 a solopreneur for the last 16 years now, not working for somebody else. And over the years, I’ve tried to
0:26:24 develop various softwares and I tried to get into the SaaS business and it was always such a pain to
0:26:30 work with developers or to try to teach myself to code. And now I know I can just go to some of these
0:26:37 AI tools and just say, here’s what I envision, go build it for me. Like, I feel like that’s valuable
0:26:42 to any business, especially the sort of, you know, small businesses and solopreneur types, which I think
0:26:44 is probably a big chunk of the people that listen to the show.
0:26:48 Yeah. You know, whether you’re in a big company or you’re solo, like, I mean, if you’re in a big
0:26:52 company, you may have a situation where you have like all these ideas, but there’s no way to express
0:26:56 those ideas or test those ideas or make them material in any way. Now with stuff like this,
0:27:00 you know, you could be like, okay, well you can’t do that during normal working hours, but at night
0:27:05 you could just like chat with AI and build like a prototype. We’re going to get closer and closer to
0:27:08 like where people have an idea and that the idea becomes real.
0:27:13 And then you can test to see if that idea was actually good or not really, really fast.
0:27:16 Where in the past, like if you had an idea, it’s like, well, okay, am I going to go spend like
0:27:21 months working on that idea? And then maybe nobody likes it. We’re going to get where you can just
0:27:24 have an idea, test it in a day, hours.
0:27:28 Have you ever read any of the Gino Wickman books, like get a grip or rocket fuel or
0:27:33 I haven’t, I haven’t traction. So he’s got this concept that every business kind of needs
0:27:38 two people to be successful. You need a visionary and you need an integrator. The visionary is the
0:27:44 person that has the big ideas that sort of sees where the company is going, pretty much has that
0:27:48 roadmap and vision of where they want to get it, but they’re not necessarily the person who’s going
0:27:53 to get into the weeds and write the code and build the website and, you know, do all the outreach.
0:27:58 They’re not the like doer in the company. So the company needs an integrator also. And an integrator
0:28:03 is that person that’s going to either go in, you know, hire and find the people that you outsource
0:28:08 to, or they’re going to be the person that develops the thing themselves, or, you know, they’re the ones
0:28:14 that gets into the weeds and sort of does the building. And to me, this really opens up the
0:28:18 possibility for like everybody to be that visionary, right? It really opens up the possibility to like,
0:28:25 I have this idea of where I want to go and I don’t need an integrator anymore. AI can be my
0:28:29 integrator, which is good and bad because everyone’s got a bunch of ideas and they’re not all good.
0:28:33 Yeah. And it’s also not great if you’re not a visionary type person, right? If you’re not an
0:28:39 idea person, right? That’s where I think there might be some struggle in the future with this AI era that
0:28:45 we’re entering into. I think this AI era that we’re entering is really, really, really good for the
0:28:51 visionary type people, but it could be really, really sort of catastrophic for the integrator type
0:28:56 person because the integrator type person are the ones that probably have to worry about, you know, what am I
0:29:01 going to do like with my skillset that I’ve built now that AI can do it. But if you’re that visionary, you’re
0:29:06 that idea person, you’re that thinker, the era that we’re entering with all of these AI tools, like best time
0:29:08 ever, best time ever for that person.
0:29:14 Yes. Yeah. Seriously. Like, like, I’ve dreamed of this my entire life. I’ve always had all these great
0:29:18 ideas. Sometimes the timing was wrong, you know, and I’ve always been like lightly technical and had to
0:29:23 deal with tons of engineers and designers and then raising capital for that. And just to think that now
0:29:30 any idea you have, you know, I kind of wish I was like, you know, 18 again, like right now, like really
0:29:36 wish I was 18. Because like, the fact that now or very soon, any idea you have, you can just create it and have
0:29:40 some level of control over the whole process. Like I said, like, you know, there’s no disconnect there. Like you
0:29:44 got an idea and then you deal with a designer and engineer. No, you can be part of that entire process to make
0:29:49 sure you actually create the thing that you envisioned. Yeah, yeah. So exciting. I’m kind of glad all this stuff
0:29:54 started to happen after I already had a wife and kids, because otherwise, I’d probably just be obsessed and never
0:29:59 leave my house. You wouldn’t have the wife and kids. I’m super happy to have my wife and my son. So yeah, I
0:30:03 wouldn’t change anything like the whole eight, wish I was 18. No, I don’t actually wish.
0:30:11 So right now we’re on the sixth step. It’s doing the CSS. All right. So just move to the seventh step. Now. Now it’s
0:30:17 just testing. So we should have something to show for this in, you know, any minute now. It actually says deploy
0:30:23 landing page and share with users. So it may actually deploy it for us. I’m curious if it’s going to like give us a link
0:30:28 that’s actually usable, or if like, it’s going to give us code that we need to go and like deploy somewhere else.
0:30:32 Yeah. Next step is it goes and markets it for you, sends you some money.
0:30:41 Yeah. Yeah. Yeah. Yeah. I mean, quite honestly, I bet you can say, go and create some Google ads for me.
0:30:47 So one thing about Manus though, is it’s not going to log into sites on your behalf. So let’s say you
0:30:53 ask it to go check your email for you and like reply to the 10 most recent emails and save them as a draft,
0:30:57 right? You can tell it to do that. But when it gets to the spot where it needs to log into Google,
0:31:02 it’s going to say, I need to hand this off to you, right? So it’ll hand it off to you.
0:31:07 You then go and enter your login details and then hand it back. And then after you enter your login details,
0:31:13 it takes it over again and starts working again for you. So you can’t like plug in your username and password
0:31:16 and have it go and log in on your behalf.
0:31:22 All right. It looks like we’ve got something here. It’s still finishing the test and validate landing
0:31:27 page, but I’m actually seeing a landing page on the screen. It called the website step style,
0:31:29 which is actually not a bad name, honestly.
0:31:34 Okay. Yeah. All right. So we’re getting a preview of this site, but I don’t have like a,
0:31:40 an actual like a link or place to check it out myself. It looks decent just from the previews we’re
0:31:44 seeing. Yeah. I mean, it’s not amazing, but it’s pretty standard. I mean, it’s, you know,
0:31:51 it doesn’t have like, you know, parallaxing background images and animations when you hover
0:31:56 and stuff, but it’s like a pretty standard looking e-commerce site. But I mean, here’s the thing though.
0:32:02 It went and found all the images. It wrote all of the copy for us. It found all of the prices for us.
0:32:07 It did the research on the popular shoes for us. Like the design is, you know, it’s standard,
0:32:11 right? Yeah. But if you look at all the other stuff that came before it, that’s the more impressive
0:32:18 part to me. We know that Claude and V zero and replet and some of these tools can do, you know,
0:32:24 okay, pretty decent looking designs. It’s all the stuff before the design that I think is impressive
0:32:27 now. Yeah. We didn’t really provide it much either, right? Like if you were doing this actual,
0:32:31 you know, a business use case of like building landing pages, you’d probably be providing an asset,
0:32:36 giving a clear list of like, here’s the kind of landing pages I want to create, maybe giving
0:32:40 it some resources or like examples. If you did that, it probably would be able to make like
0:32:44 tons of landing pages for you. And they’d probably be good enough to test if I had to guess.
0:32:49 For sure. So that seems like that’s really going to change marketing, like being a marketing manager
0:32:53 or whatever, like you have an idea for like, you know, marketing campaigns or like landing pages
0:32:57 you want to have created. Like now that person really could just do the entire thing in theory,
0:33:01 right? Like they come up with the idea and then just have AI create all the landing pages.
0:33:05 Eventually have AI help create the ads. Maybe it can’t do that well right now,
0:33:06 but I’m sure that’ll be coming.
0:33:12 You can actually have it log into like your Google ads account for you and then it hands it off.
0:33:17 You plug in your username and password and give it back. And it can probably go and write the ads
0:33:22 for you. I don’t see why it wouldn’t do that right now. You probably wouldn’t want it to just deploy the
0:33:26 ads without you double checking them first, but it would probably go and like fill out all the
0:33:30 details for you. Yeah. That’s probably a good experiment for people in business to be running is like
0:33:34 doing it. Like I’m a very limited amount. Like it’s very small budgets, like a hundred dollars or
0:33:38 a thousand or whatever is small for you. You know, one thing that I think would be interesting to test
0:33:44 would be like, can it go into like Canva and create an ad for me? Right. Can it like actually
0:33:49 like manipulate and drag the images around and like actually design an ad for me? That’s like a banner
0:33:54 ad or something. Why don’t we try that? I bet it can’t though. If I had to guess. Yeah. Maybe in one of
0:33:59 our future episodes. Right. Right. One thing I’d love to see Manus do is have some sort of like
0:34:03 environment variables that you can plug in, right? Where you can actually give it usernames and
0:34:08 passwords. And ideally it’s in some sort of secure location and, you know, save to your local computer
0:34:13 and maybe not save to Manus’s servers. Right. I don’t know how that would work exactly, but it would
0:34:18 be nice if it can like sort of do some of the logging in on your behalf. So it looks like it’s finishing
0:34:23 up the last step now. I mean, this is huge progress. I mean, like think about baby AGI. It would come up
0:34:26 with that list of things to do and it would go off and do some research. It would do some basic
0:34:29 research. Wasn’t that great. And then it would probably go to the next step and then just get
0:34:33 stuck. Like, okay, let’s go start making the website. And it was just like, how do you do that?
0:34:35 Yeah. Yeah. Yeah. It would get stuck very quickly.
0:34:40 Okay. So it’s asking me a question now. So it says Manus is attempting to deploy the service
0:34:45 located at blah, blah, blah to the public network. Please confirm if you agree by clicking the button,
0:34:50 allow Manus to deploy the application to the public network. Um, sure. I don’t know where
0:34:55 the public network is located, but it’s nothing sensitive. So I’m not too worried about it.
0:35:01 It’s a fake shoe website and website clones are going to get really bad. Yeah. Yeah. I mean,
0:35:06 you look at like blue sky, right? And blue sky just kind of looks like a clone of Twitter already.
0:35:10 Yeah. Yeah. Definitely. You’re going to see a lot more like that. I think where people are like,
0:35:13 we’re launching a new social media and all it is, is a clone of something else,
0:35:16 but they change the color scheme. Yeah. And that’s where you learn, like, you know,
0:35:21 network effects really matter and distribution really matters in those things. That’s like
0:35:25 actually one of the core things to them is not even the technology. It’s the, the users and the
0:35:31 distribution. Yeah. All right. So now it’s editing the to-do file. All of the task progress are checked
0:35:36 off. It showed a URL here a second ago, but then it disappeared. All right. So to successfully
0:35:42 deploy the shoe website landing page, ready to share final results with the user. All right.
0:35:48 Share them. Let’s see. Aha. Okay. So here’s what it says. The landing page includes all the elements
0:35:54 you requested, social proof, research on best shoes, images of shoes, customer testimonials,
0:36:01 view all files in this session. Okay. So it gave me all of the files. So if we did want to deploy these,
0:36:06 what would be the best way to deploy them? Would you say like, so that other people can
0:36:11 actually access this? So here’s what I would probably do. I would probably load it into a
0:36:17 GitHub repository. Yeah. And then attach my GitHub repository to Vercel. Yeah. That’s what I was going
0:36:20 to say. Probably not the easiest way though. Like for like most people, there’s, there’s gotta be some
0:36:25 other way. I know when you use replets agent, it’s very easy, but I don’t know if you’re not using
0:36:29 their agent, how easy it is just to push something alive like this. Yeah. That might be a little too in
0:36:35 the weeds for this one, but I would imagine if you wanted to, you could tell Manus to push this to
0:36:41 GitHub for me and then set up a Vercel account, but you would have to keep on taking control every time
0:36:46 it was asking for login details, but it would walk you through it. At least it would be like, here’s
0:36:51 the next step. I need you to log in for me. You log in, you hand it back. And then here’s the next
0:36:55 step. I need you to log in for me. You log in and then hand it back. And then it would actually do
0:37:00 the process of setting up GitHub, setting up for cell and pushing it all. So it says Manus has
0:37:06 completed the current task and is ready for new requests or revisions. So it didn’t actually seem
0:37:13 to give me a live link where I can go and view it, but it did give me all of the files here. All right.
0:37:20 So links, no items, code files, images, documents, all. So we do have all of the files here with the
0:37:27 research notes, testimonials, stats, index.html style script to do. Oh, here it is. I’ve completed
0:37:34 your shoe website landing page open. Okay. So it actually put it online for us. It’s at this URL.
0:37:39 I don’t know if that’s like temporary or if it’s like fully deployed for us and we can just like
0:37:47 change the domain name, but here’s the website step style shop collection, explore categories. So if I
0:37:52 click explore categories, it sort of auto scrolls down the page for me, the numbers speak for
0:37:59 themselves. It’s got the, you know, global footwear market 495.7 billion. So it actually has some like
0:38:05 research stats on the page about the shoe market featured collection. This is based on our research,
0:38:11 right? These are trending shoes, the retro runner sneaker, Python print pumps, soft leather loafers,
0:38:16 performance runners. And it gave you all the research files too, right? It gave you like all the stuff
0:38:20 it collected. And so that’s crazy because you could do this same kind of thing I’ve been using for
0:38:26 projects where you could then feed that content into an LM and like kind of have a conversation about it
0:38:30 and like, okay, let’s talk about the research and then how do we want to change things based on that
0:38:34 research, right? Yeah. And then you could start iterating on it. Use this like the baseline for the
0:38:39 starting point. I mean, none of the links work. So it seems to be just like a single page without,
0:38:45 you know, any internal links yet. Yeah. But you’ve got the bones of a front end landing page.
0:38:51 We’ve got the science of comfort with some research text here. We’ve got some fake testimonials that
0:38:56 it made up for us and added stock images. I mean, it actually looks pretty clean. I mean,
0:39:00 it’s not like the best looking website ever, but it’s clean. You wouldn’t look at this and be like,
0:39:03 oh, this site looks like crap. It’d be a decent starting point.
0:39:06 Yeah. You take something like this into, you know, cursor or something like that, and then
0:39:10 have a conversation with it about what you wanted to change. And then, you know, probably after a day,
0:39:14 it’d be pretty good. Yeah. And I mean, Hey, if you wanted to create just like a landing page that
0:39:18 collects emails, you know, or is in the marketing world, they call it a squeeze page, right? If you
0:39:22 wanted to create a squeeze page, that’s like a single landing page that collects emails,
0:39:29 this would be a pretty easy way to do it. So yeah, that’s Manus. I mean, I would say after seeing it,
0:39:33 do you have any like, you know, additional thoughts, but I feel like we pretty much shared most of our
0:39:35 thoughts as it was processing.
0:39:40 Yeah. I mean, I guess the only thing is I find it super interesting. I think that probably though,
0:39:44 if I had to guess, everyone’s going to copy it, like how this is working. It’s not clear that
0:39:47 Manus is going to be the winner in this space. I’d be shocked if open the eye and Claude and others
0:39:52 don’t do the same. Cause I think agents is going to be a core part of their business.
0:39:57 Oh yeah. A hundred percent. Well, you know, one of the things I mentioned was talking about some
0:40:01 alternatives, right? Just because there is a 2 million person waiting list and they’re only letting a
0:40:06 handful of people in every day. Well, this week, the week we’re recording the week before you’re
0:40:11 actually listening to this open AI just released some new features inside their API. Here’s the
0:40:17 blog post from March 11th, new tools for building agents. So in their API, they actually gave people
0:40:22 access to sort of create these like agentic features in the apps that they’re building directly
0:40:28 through the API. So agents easily configurable LLMs with clear instructions and built-in tools,
0:40:33 handoffs, intelligently transfer control between agents, guardrails, configure safety checks,
0:40:37 tracing and observability. Yeah. One of the big parts of the release. Well, one of them was
0:40:42 they updated computer use. Apparently the computer use is doing way better on benchmarks. I haven’t had
0:40:46 a chance to test it yet. So that’s one. I remember it went up by like 15 or 20% on the benchmarks,
0:40:51 I think. Yeah. Another one is like, uh, the agents collaborating with one another. Yes. Like they
0:40:56 apparently really improved upon that. And that’s part of this release. Yeah. But most of it seems to be
0:41:01 like through the API, right? I don’t know if there’s like a front end UI where you can just go and play
0:41:07 with this. So, yeah, I mean, as far as like alternatives go, you do have the chat GPT operator.
0:41:14 The open AI operator agent is currently available on the pro plan for $200 a month. However, there are
0:41:19 plans to eventually make it accessible to users on the $20 a month plan. So I wouldn’t really say this
0:41:23 is a great alternative because you got to pay 200 bucks a month to get access to operator, but this
0:41:29 will do some similar stuff. And then there’s also this browser use tool, which is an open source
0:41:34 tool that you can use. It’s actually a Chrome extension that you can use. They do have a free
0:41:39 open source version. I’ve played with it. This was the tool that I used in a previous live stream
0:41:43 that I did where I asked it to go to Amazon for me and buy me some toilet paper. And it went and added
0:41:48 $600 worth of toilet paper to my cart. And then I canceled it before it actually placed the order.
0:41:53 So there is that you can go play with it, but browser use is actually what Manus is using under the hood.
0:41:59 I just think Manus is a lot better at sort of giving it good instructions where just giving it broad
0:42:04 instructions, like go order toilet paper is not clear enough, but Manus, I think using
0:42:09 the underlining like Claude or Quinn models is better at making sure it won’t go do stuff like
0:42:14 that. There’s also this other one I came across called highlight AI. I haven’t played with this
0:42:18 one yet myself, but they say everything you love about cursor across every app on your computer.
0:42:24 So if you look at the app, it’s basically a tab that opens up on your computer and you can ask it to do
0:42:29 stuff and it will open up different apps and run programs for you and stuff like that. Again,
0:42:33 not what I’ve tested, just wanted to show it else is available out there. And then there’s this one
0:42:41 called pig at pig.dev and it says operate windows apps with AI pig makes it easy to build complex
0:42:46 automations powered by AI. And this one actually will take control of your mouse for you. So it’s not
0:42:51 using like a virtual browser. So I don’t know how much I would trust this one yet. Cause it’s literally
0:42:57 taking control of your own computer, not a virtual computer. This one though, you do install and I
0:43:03 believe it’s open source. I’m not a hundred percent sure on it. I think they could have picked a better
0:43:07 name for something that you’re going to trust with your computer to like run your computer calling this
0:43:13 pig. Yeah. So it does say a pig lit is fully open source and you can check out the code on GitHub.
0:43:20 So pig lit is, I guess the open source version of this. Again, this one actually takes control of
0:43:29 your computer where computer use and operator and menace all seem to use like a virtual desktop that
0:43:35 it uses that it will go and take the actions inside of. I think for now, the virtual desktop is the
0:43:39 better way to go. Like when these things get really good and you can actually trust what it’s doing.
0:43:44 I think giving it access to your computer will probably make sense, but for now, I think I
0:43:49 would stick with the ones where the virtual sandbox. Yeah, I agree. I’m too scared to put pig on my
0:43:54 computer and a, it’s a tool called pig. Like you mentioned like that just doesn’t make me too
0:44:01 confident in it, but B I don’t want an AI going and controlling my mouse and give me back my mouse.
0:44:04 I can’t do that, Matt. You know, like I don’t want that scenario.
0:44:09 We’re locked in, Matt. You told us to do this, Matt. We’re going to finish it.
0:44:17 Exactly. But anyway, that’s a sort of breakdown of the current standing of the world of AI agents.
0:44:21 Obviously, Manus is the big talk of right now. A lot of people creating some really cool stuff.
0:44:25 I’m fairly impressed with it. I wish it was a little bit faster right now. It feels a little
0:44:30 bit slower than it should be, but that could just be a symptom of so many people all trying
0:44:33 to use it at once, right? If it wasn’t so bogged down, it might go a lot faster,
0:44:38 but it is a big moment because like agents are working now. Yeah. Like there’s been hyper agents
0:44:41 and it’s like everyone was saying like, this is going to be the year of agents. They are working
0:44:47 now and we’re only in March. Yeah. It is officially the year of AI agents. They are here. Yes. And we’ve
0:44:51 got like over nine months of more improvements. So I mean, like by the end of the year, agents are going
0:44:56 to be amazing. So anyone business, I think that’s what you should be thinking about is like be testing
0:45:00 these things out to like understand how they work and then try your best to extrapolate and like think
0:45:04 about over the next year. What does it mean when this thing is like five to 10 times better? That’s
0:45:08 most likely going to happen. And how does that impact your business? And how can you be using it?
0:45:14 Because that’s what I’m trying to think about daily. Absolutely. Absolutely. Well, I think that about
0:45:20 covers everything we can kind of possibly talk about around Manus and the current state of AI agents.
0:45:23 So we can go ahead and wrap this one up here. If you like episodes like this, make sure you’re
0:45:28 subscribed either on YouTube or wherever you listen to podcasts. We have a lot more stuff like this
0:45:32 coming next week. I’m going to share my whole breakdown of how I’ve been rebuilding future
0:45:36 tools. We’re going to talk about some other really cool apps that have been coded with AI recently and
0:45:41 break them down a little bit better. So that’ll be a fun episode that we’re releasing next week. So
0:45:45 again, don’t forget to subscribe and thank you so much for tuning in. We’ll see you in the next one.
0:45:47 Thank you.
0:46:03 I’ll see you in the next one.
0:00:11 right now, agents are all of the rage. We are in the era of AI agents. And right now,
0:00:15 the biggest agent that everybody is talking about is Manus. So in this episode, we’re
0:00:19 going to break down what is Manus. We’re going to show you what other people have created
0:00:24 with Manus. We’re going to use Manus ourselves and actually build a really cool website with
0:00:28 it. You’ll see what it built before the end of this episode. We’re going to break down
0:00:33 how Manus was made, the bigger implications, how you can use it in your business. This
0:00:38 episode is jam packed with AI agent goodness and how it’s going to impact your life and
0:00:42 your business. So get ready. You’re going to be in for a fun ride. Let’s get right into
0:00:50 it. HubSpot just dropped their 2025 marketing trends report, and you’re going to want to
0:00:57 see what they found. Visual content is delivering 21% more ROI. Small influencers are building
0:01:04 45% more trust and AI is changing how fast we can create. And the best part, it’s not
0:01:10 just another report full of stats. It’s a game plan you can actually use. There are frameworks,
0:01:15 AI guides to help you do more with less and real case studies and playbooks from teams crushing
0:01:22 it right now. Want to nail marketing in 2025? Go to click HubSpot.com slash marketing to download
0:01:32 download it for free. So where should we start? Maybe we should just start showing off some of
0:01:36 the stuff that other people have made with Manus. Yeah, I think there’s been so many crazy posts on
0:01:40 X like showing how people are using it. I think let’s just start there and show what people are
0:01:46 doing with it. You and I have both played with it a little bit already, right? Like I shared a screenshot
0:01:53 on my X account of me trying to play with it. And in this tweet, I basically gave it the prompt to go
0:01:59 research Manus because I want to make a video about Manus. So go research Manus. What is it good at?
0:02:05 What is his limitations? What’s the competition? You know, just give me as much deep dive info as you
0:02:13 can on Manus. And it broke right during the portion where it was investigating the limitations of Manus.
0:02:20 So literally the spot that it decided to stop working was when it was starting to deep dive on
0:02:25 its own limitations. Yeah, that makes sense. As intended. But anyway, I know you’ve tried to play
0:02:30 with it as well and sort of ran into some… No, I just tried to play with it a little bit. Like it took
0:02:35 forever to get access. You know, I had to message a bunch of people on X to get access. Then I finally got
0:02:39 access and I don’t know if it was just the time zone difference in Japan. Maybe they were doing
0:02:44 some maintenance or something. But when I tried it, like it wasn’t working. So I haven’t actually got
0:02:48 to use it yet. I tried it one time, didn’t work. And so I was like, well, you know. Yeah, yeah, yeah.
0:02:53 Well, I think what’s happening is that Manus is insanely overloaded. They actually tweeted this
0:03:01 on the day that we’re recording. Seven days, two million people on the wait list. So yeah, I think,
0:03:05 you know, they’ve been letting a lot of people in every day. I think they’re letting like,
0:03:09 I don’t know, 50 to 100 people in a day or something like that. And I think they’re already
0:03:13 overloaded. So I think that’s what we’re seeing right now is just so many people trying to use
0:03:19 it at once. Even that one that I was just showing you here where it says investigate and document
0:03:24 Manus AI technologies limitations. And that’s where it actually stopped working. If we look over here on
0:03:32 the left, it says high system load has caused an internal server error. So it didn’t like get messed up
0:03:36 because it didn’t want to, you know, share its own limitations. It got messed up because
0:03:42 at that point in its own processing was when the servers got overloaded and it just sort of stopped
0:03:48 working. But yeah, let’s take a look at what some other people have made here. So this one is from AK
0:03:57 and he actually got it to make this like instant run game, very similar style to that, like levels IO
0:04:03 flight game, a lot more basic, obviously. Yeah. But if you’re listening on audio, it’s basically a
0:04:09 character just like running in a straight line and it’s like a 3d world, but everything’s sort of like
0:04:14 basic shapes. It’s like squares and cones and stuff. It’s very, very basic looking, but it’s just a
0:04:18 character running and obstacles getting in its way. And it’s obviously trying to dodge the
0:04:23 obstacles and like grab power ups and stuff. But that’s just from a single prompt, right? So like
0:04:30 it went off and did everything. Yeah. Literally AK says the prompt was make a three JS endless runner
0:04:36 game. Yeah. So one, two, three, four, five, six word prompt made that. Yeah. And he specified three
0:04:41 GS and it’s kind of interesting. Like I’ve played with three GS for like years and it just never took off
0:04:45 really. And then it’s crazy now that it’s taking off because of AI. Well, give me some more context.
0:04:50 What is three JS? I’m a total non-coder. I think it’s like a JavaScript library for like 3d stuff.
0:04:55 Oh, okay. So here’s one from our buddy, uh, Bilal here. He says I tested Manus AI. It’s the coolest
0:05:01 thing I’ve experienced to a truly autonomous AI agent. I can’t wait till this thing can use desktop acts
0:05:08 like premiere and Photoshop. It low key feels like baby AGI. That was actually my first thought. Like when I
0:05:14 actually was using Manus the first time, it felt very similar to baby AGI. We’ll show it off in a
0:05:19 minute if it works, hopefully. And what it’ll do is it actually builds out a plan for you. And then
0:05:24 once it builds out the plan, it starts back from the top and then tries to execute each step in the
0:05:30 plan, which feels very similar to what baby AGI was doing in the past. I mean, baby AGI was a great
0:05:34 concept when Yohei created it, right? It was just, I think the models were not good enough to actually
0:05:38 execute on those plans. It’s exciting to see that now they’re finally getting good enough to actually
0:05:44 start executing on, on the plans they generate. Yeah. Yeah. For sure. Yeah. I mean, when baby AGI
0:05:48 came out, it would get stuck in these loops. Yeah. It would like go to step three and then for some
0:05:52 reason, just like keep on searching the web and never get past step three. Yeah. This one actually
0:05:58 seems to, you know, actually follow through and finish, but here’s the prompt he gave. Can you open
0:06:04 up a 3d view inside Google maps of Austin, Texas? And let’s find the most compelling photos to take
0:06:10 of the downtown skyline. It’s very important that you change the map perspective so we can get oblique
0:06:18 3d views versus a top down one. I want you to fly around surveying downtown from 3d vantage points
0:06:24 in Google maps and tell me your favorite location slash photos. So that’s the prompt, which is,
0:06:30 I mean, Bo Aval uses big words. I love him, but he does, but you can see here very similar to baby
0:06:37 AGI. It went and created this task list for the agent to follow, but you can see it says research drone
0:06:41 regulations in Austin, search for potential drone locations near downtown Austin, create a short list
0:06:48 of locations, scout locations using Google maps, 3d view, evaluate locations for drone flight suitability,
0:06:54 complete top three recommendations, prepare final report with recommendations, deliver results,
0:06:59 to user. So that’s the sort of task list that it gave. And then you can see as he scrubs through,
0:07:05 it goes through, it does all these tasks. We can see here, it’s actually looking on Google maps for him,
0:07:12 trying to find locations, and then eventually gives him top three recommendations, why it’s recommended.
0:07:17 So, I mean, that’s pretty wild that it can even do that in my opinion.
0:07:24 And then we’ve got this here. This is a thread from Elsine over on X. And this is a thread of a whole bunch
0:07:31 of different shares that he found. So the first one, he had it clone the Apple website. This one seemed like
0:07:37 a simple task to me, but we can see here, it literally made like a exact duplicate of the Apple
0:07:38 website here.
0:07:42 Wait, did it take a screenshot of the Apple website? Is that how it was generating the website?
0:07:42 I’m pretty sure.
0:07:46 It looks like it actually like took screenshots of some of the elements, right? Just like pasted them
0:07:47 in, which is hilarious.
0:07:52 But then we’ve got an analysis of Tesla stock where it went and did a whole bunch of research on Tesla.
0:07:58 I don’t really feel like you need Manus for this. I actually did this exact one. I had it go and
0:08:04 analyze Tesla stock for me using ChatGPT’s deep research. And it was really, really, really good.
0:08:09 So I feel like Manus is almost overkill just to do this research here.
0:08:14 Yeah, you can use that, Grok, Perplexity. They’re all great for use cases like that already.
0:08:22 Yeah. Plan a family trip again. Do you really need like an AI agent to plan the family trip? I mean,
0:08:27 an agent like Deep Research, you know, from ChatGPT or from Google or from Perplexity,
0:08:32 like those deep research agents are really good at planning family trips. I don’t really feel like
0:08:37 you need Manus to do that. The ones that I think have impressed me more have been more these like
0:08:43 visual ones. Like here’s this game that somebody created again with 3JS. Yeah, we can see the tasks
0:08:48 that they did to make this game create project structure for 3JS playing games, set up 3JS
0:08:53 rendering environment, design and create plane 3D model, simplify left, right plane controls,
0:08:57 fix plane shooting, test, improve plane movement and controls, finalize game implementation,
0:09:01 share updated game version with user. And then on the right side of the screen,
0:09:06 we can actually see the plane game kind of look similar to, you know, Peter Level’s game and that
0:09:13 game we saw from AK similar style with where it’s real basic shapes, but it did it with a single
0:09:20 prompt. Website SEO optimization. Again, not really a great use case for Manus in my opinion.
0:09:26 Double agent productivity. Somebody actually showing they were driving their Tesla and they had Manus
0:09:32 running on the screen in their Tesla doing some autonomous stuff. This one was actually shared by
0:09:36 Manus. So it’s a little bit more legit. Double agent productivity while Tesla’s full self-driving
0:09:42 takes a power user to his coffee meeting. Manus preps him with all the key talking points. So he’s on
0:09:48 his way to a meeting while Manus is rounding up talking points for him to actually talk about at the
0:09:54 meeting. He’s going to, that’s crazy. Like you had the glasses, like just telling you, like, what you’re,
0:09:59 what am I talking points? How should I respond in this negotiation? Like here’s the parameters of like
0:10:03 how I should negotiate this deal. Like, Oh, I should ask for more right now. Look, look at how he’s
0:10:08 responding. You know? Yeah. Yeah. Yeah. Well, I mean, with the project Astra glasses that I actually
0:10:12 demoed when I was at deep mind in London, like you’re going to be able to do stuff like that. It’s hearing the
0:10:16 conversation around you and we’ll be able to actually like put little text on the screen in
0:10:20 front of you that you can read off of. And one of the cool features of that is you can almost use it
0:10:24 like a teleprompter. I can see getting those glasses and like making a YouTube video and like literally
0:10:28 reading off a script the whole time, but trying to hide the fact I’m reading off a script because it’s
0:10:32 like all right inside of my glasses. In fact, I might do that for future HubSpot sponsorships.
0:10:39 We’ll see. Let’s see. Create an amazing animation using P5JS and it created this like colorful
0:10:45 animation. So those are some of the things that other people have created with it really quickly.
0:10:50 I think we should talk about like how Manus was actually developed because this was sort of a point
0:10:56 that you brought up that you saw from Gary Tan. Yeah. But, you know, people were complaining going like,
0:11:04 Hey, um, I just simply asked Manus to give me the files at opt slash Manus and it just gave it to me.
0:11:12 It’s Claude Sonnet with 29 tools using multi-agent and browser use. Yeah. So basically Manus is just
0:11:17 using Claude and this browser use, which I believe is open source. And it was kind of funny people when
0:11:23 they discovered that some people started calling it like a fraud, like, no, it was like, I get what
0:11:27 they’re saying. Like, yeah, they even like created some amazing proprietary model or whatever. Maybe they
0:11:31 have like, it’s kind of like, you know, one on top of Claude, but, but still like, I mean, they’re
0:11:35 combining it together in a very unique way. And like, you know, to me, that’s the same as saying
0:11:40 that, Oh, a cursor or windsurf is a fraud because they didn’t make their own model. Yeah. Yeah.
0:11:44 I mean, perplexity actually has its own LLMs now. Like they have their own fine-tuned, um,
0:11:49 llama models, but yeah, it’s like calling one of those tools a fraud. Cause you didn’t train the
0:11:55 model yourself. Like who cares? It doesn’t make any sense. But then, you know, this guy
0:12:00 peak here, he’s the CEO. He said, Hey, I’m peak from Manus AI. Actually, it’s not that complicated.
0:12:06 The sandbox is directly accessible to each user. Each session has its own sandbox, completely isolated
0:12:11 from other sessions. Users can enter the sandbox directly through Manus’s interface. The code in
0:12:16 the sandbox is only used to receive commands from agents. So it’s only lightly obfuscated. The tool
0:12:21 design isn’t a secret basically saying like, we’re not trying to keep anything a secret, right? Right.
0:12:26 When you ask it a question, it’s going to figure out which models to use. I guess it uses both Sonnet
0:12:32 and Quinn. So based on your prompt, it sort of decides which models to use for the various steps.
0:12:38 And then it does use this open source browser use tool, which is a pre-created open source tool that
0:12:43 will actually go and use your browser. Although it’s not using your browser. When you do use Manus,
0:12:47 it sort of opens up a virtual computer and uses this browser use tool in a virtual computer.
0:12:53 We’ll be right back to the next wave. But first I want to tell you about another podcast I know
0:12:58 you’re going to love. It’s called Marketing Against the Grain, hosted by Kip Bodner and Kieran Flanagan.
0:13:04 It’s brought to you by the HubSpot Podcast Network, the audio destination for business professionals.
0:13:08 If you want to know what’s happening now in marketing, what’s coming, and how you can lead the
0:13:12 way, this is the podcast you want to check out. They recently did a great episode where they show you
0:13:16 how you can integrate AI into the workplace. Listen to Marketing Against the Grain wherever
0:13:18 you get your podcasts.
0:13:26 Yeah, I think it is interesting that this year, like all the top American AI companies have been
0:13:32 trying to push out solutions with agents and they haven’t really got much attention yet. And then
0:13:35 this Chinese, which is not open source. Yeah, they said it’s gonna be open source, but this Chinese
0:13:40 company comes out and just combines all these tools in an interesting way and they make it happen.
0:13:45 So I think that’s gonna be a dynamic we’re going to continue to see. I should show the Gary Tan tweet.
0:13:50 Yeah, yeah. So here’s what the tweet says. It says, if real, this actually emphasizes the models
0:13:56 are plenty smart already and all the alpha is in custom prompting, tool use, and clever workflow
0:14:01 and evals. Yeah, I think that is probably where things are at. Even though OpenAI has got a great
0:14:05 model and Claude and everyone else, man has figured out how to combine these together in an interesting
0:14:10 way first, you know? Yeah. Which is exciting because it feels like all these models, they’re just waiting
0:14:13 for people to like build things on top of them, like new things and try new things.
0:14:19 Yeah. Yeah. Again, I don’t think it’s a bad thing that they went and used existing large language
0:14:24 models and existing open source tools. They’ve just combined them in a way that other tools haven’t
0:14:29 yet because they went and combined them in a way that others haven’t. They made something really
0:14:36 valuable and useful to people. And like Gary says here, it just shows these models are a lot better than
0:14:40 people are giving them credit for. Right. Right. It shows that these models are already capable of
0:14:45 going and just handling work for you. You just need to sort of have the right workflow and the right
0:14:50 combination of large language model and tool use to get done what you want to get done.
0:14:56 Yeah. I do wonder like, why did Claude not do this first? I wonder if it’s out of restraint kind
0:15:00 of thing, or they just didn’t come up with it. You know, I’m not sure which one it is.
0:15:04 They had, what was it called? Claude computer use, right? Yeah. Or something like that.
0:15:09 Claude was kind of the first one to push something like that out. Yeah. But it was such a pain in the
0:15:14 butt to actually use. Right. You had to install Docker on your computer and then you had to run a whole
0:15:19 bunch of like terminal commands and install a bunch of stuff. And like, even if you’re not that technical,
0:15:24 it’s pretty easy to follow the directions. It was just a lot of steps that I think scared a lot of
0:15:28 people off. Right. So nobody ever really used Claude’s computer use because it was like
0:15:34 too many steps. Yeah. And then ChatGPT came out with their operator, which, you know, simplified
0:15:38 things a little bit. Oh yeah, that’s true. And it’s actually doing something similar to this. It’s got
0:15:42 like a sandbox where like, instead of you having to put it on your own computer, they’ve got like a
0:15:48 virtual machine running. Yeah. But I guess, you know, Claude 3.7 is just, you know, that was,
0:15:54 that was the magic, right? That was the one that made it work better. You know, operator launched a few
0:15:58 weeks ago, maybe a month ago. I don’t remember exactly when, but it’s been out for, you know,
0:16:03 three, four or five weeks somewhere in that range. And it was all hyped up the week that it came out.
0:16:06 And then I feel like nobody’s been talking about it ever since. Right. It’s just sort of like died
0:16:12 off. So it must not have impressed people that much. And now we have Manus and like, everybody can’t
0:16:17 shut up about it now. And it’s like, well, it kind of is similar to what we already have, but it just
0:16:24 seems to work better. It just seems like this combination of Claude 3.7 plus browser use, plus,
0:16:31 you know, occasionally using Quinn as well was like that right mix to make it feel valuable and make it
0:16:36 feel like it’s really working for people. Yeah. I mean, it does feel like open AI, you know, they’re
0:16:40 trying to do so many different things. Yeah. Like early on, they were doing AIR too, which apparently
0:16:44 they have a big update coming or whatever, but you know, mid journey and Leonardo and other people are
0:16:48 like totally dominated that space. And so it feels like maybe it’s the same kind of thing where,
0:16:53 you know, the core model companies, they’re trying to do so many different things. And so a company can focus on
0:16:57 one thing like, okay, let’s make sure that the, uh, the AI models know how to use a computer really
0:17:02 well or use tools very well. Uh, they can still win by just having like really good focus.
0:17:08 Yeah, no, absolutely. Well, let’s go ahead and try out Manus. Yeah. Hopefully it works and it’s not
0:17:12 overloaded. And then after that, I have like a few alternatives that I’ve come across. Like we
0:17:17 mentioned, there’s 2 million people on the waiting list. So I thought it would be helpful to share some
0:17:22 other options that you can go and try out if you can’t get into Manus, but what do you want to test
0:17:29 here? Um, I mean, the easiest one would be like a website or a game. I think as of right now, yeah,
0:17:33 yeah. A website would be cool. I mean, that’d be good for like a business audience. I think a landing
0:17:38 marketing page would be probably, you know, a value for business people because you’re always making
0:17:42 landing pages and then pushing ads to push people to those landing pages to test them.
0:17:50 agents make that dramatically easier to do. Yeah. Create a marketing landing page for a website that
0:18:01 sells. What should we sell? Shoes? I don’t know. It sells shoes include social proof research on the
0:18:07 best shoes. I want to give it some extra details so it’s not just, you know, going off and cloning a
0:18:18 website. Yeah. Let’s see what else, uh, images of shoes and customer testimonials with headshots.
0:18:25 Let’s go use stock photos for now. Let’s see what it does. So it’s thinking if you already had images,
0:18:28 could you have provided that to it? Uh, there is an upload button. You can see there’s like a little
0:18:33 attach button down here so you can’t attach other stuff to it. Yeah. Let’s see. Uh, I’ll help you create
0:18:37 a marketing landing page for a shoe website. I’ll include social proof research on the best shoes,
0:18:42 images, and customer testimonials with headshots using stock photos. Let me get started on this
0:18:48 right away. So now it’s setting up a project structure, creating a to-do list, updating the
0:18:52 plan. And then I believe if I open this, yeah, we can see the plan that it mapped out for us.
0:18:57 So here’s the plan that it mapped out. Set up project structure, gather stock photos,
0:19:03 create customer testimonials, design HTML structure, add CSS, test and validate landing page functionality,
0:19:10 deploy landing page and share with user. So that’s what it’s doing. And then I can actually open this
0:19:15 up and we can actually see it taking actions in our browser right now. So what you’re seeing on the
0:19:20 right side of my screen is actually it just going hands-off. I’m not touching anything. That’s just
0:19:24 literally a video. This is the future of work. Yeah. This is literally just a video.
0:19:29 It’s watching the AI. It’s on the Vogue website. We can see every once in a while,
0:19:34 it sort of flashes bounding boxes on the screen. I think what it’s doing is sort of reading the CSS
0:19:38 and then, you know, pulling text from within certain divs or something, but it’s reading
0:19:45 Vogue.com right now, I guess about the best shoes. Okay. Hopefully it’s not just copying their website.
0:19:50 Right now it says it’s doing research. According to the task list, it says research best shoes and
0:19:55 current market trends for landing page content. So according to Manus, it’s just doing research.
0:20:01 It’s not, it doesn’t seem to be like screenshotting. It seems to be like using bounding boxes to like
0:20:07 find where the text is and then reading the text inside the box. Now it’s on a site called Birdie.
0:20:11 I mean, I assume this is not going to be perfect, but like you can see where this stuff is going,
0:20:17 you know? And even recently, I’ve already found that this has changed how I like hire contractors and
0:20:22 stuff. Like my own website, Lord.com, like I built it myself, but I was busy and I needed small tweaks.
0:20:27 I might occasionally hire that person to make tweaks to the website and stuff like that. And now I found
0:20:33 it doesn’t even make sense to hire that person anymore because like I just use Claude and just type
0:20:38 something in and it changed my entire website for me. Literally like in like five minutes, it made dramatic
0:20:42 changes to my website. I added like the next wave podcast, all that. I literally just told Claude to do all
0:20:47 that. Yeah. Yeah. Yeah. And it did all of it like five minutes and I pushed it live and didn’t have
0:20:52 to hire anyone. Yeah. Well, I mean, as you know, like I’m actually rebuilding future tools and yeah,
0:20:58 I hired a developer to help me build it. The developer kind of got really, really flaky and then started
0:21:02 telling me stuff that was originally planned was outside of scope. And it was this whole pain in the
0:21:08 butt process. I eventually ended up just like letting him go. I was out like basically 10 grand on this
0:21:12 redesign, but I went, this is just getting too insane. Yeah. It’s getting more and more expensive.
0:21:19 I’m going to cut my losses. And yeah, I mean, I’ve basically managed to completely rebuild future
0:21:25 tools with a better design, like just totally by myself using these AI tools. I think you’re going
0:21:29 to see this in like so many different areas of business. So, you know, I’ve been like doing more
0:21:35 and more game design stuff lately and been watching videos of Jonathan Blow, pretty famous and notorious
0:21:38 game designer. And he was talking about the difference between like a gigantic
0:21:44 company making a game or a small team or an individual making a game, how dramatically different
0:21:49 those things are. And he was talking about how so much is lost between like a designer or a business
0:21:53 person or an engineer all wanting different things or having different things in their head
0:21:58 and communicating that to each other is almost impossible. Usually there’s usually a lot of things
0:22:02 that are beyond just words and it’s really hard to express those. And maybe even impossible because
0:22:05 people have different understandings of different words and what you mean.
0:22:06 Yeah. Yeah.
0:22:09 And it ends up, you know, in a different result. I think you’re going to see more and more where
0:22:13 you’re like, okay, your idea of future tools, instead of you having to work with a designer
0:22:17 and working with an engineer and work with all these different people, you know what you want.
0:22:17 Yeah. Yeah.
0:22:22 And also you’ll know very quickly if you like what you’ve produced and if you want to change
0:22:26 it versus having to talk to all these other people who will never actually know what’s
0:22:27 inside your head.
0:22:27 Yeah.
0:22:30 What you want. And I think you’re going to see this in so many different areas where people are
0:22:34 going to be able to create amazing new things that, you know, whole committees of people
0:22:39 can’t really create, but you and your AI partner will be able to.
0:22:44 Somebody was asking me on X, like, but don’t you get like really frustrated when you’re trying
0:22:47 to code something and you run into a bug and you can’t get past it and it takes you forever
0:22:52 to get past that bug. And I went, yeah, I mean, that’s definitely frustrating, but it’s
0:22:57 not nearly as frustrating as having something that I don’t like on my website, going to my designer
0:23:00 or my coder asking him to fix it and not hearing back for seven days.
0:23:01 Yeah.
0:23:06 Because if I run into a bug while I’m coding and I’m struggling to get it fixed, it’s usually
0:23:11 like between 20 minutes to maybe like two hours of me going back and forth until it’s fixed.
0:23:15 If I was working with a developer, it could be two freaking weeks before I hear back from
0:23:15 him.
0:23:19 Plus an invoice because it’s out of scope, you know, the developers too, it’s like dealing
0:23:23 with a mechanic. Like if you’re like a hardcore developer, you have no idea what they’re saying
0:23:26 and if it’s true or not. And that’s actually another thing Jonathan Blow was talking about.
0:23:29 He’s like, he’s been that same dynamic where like a designer used to tell him something to
0:23:33 code and then he like disagreed with it, but he would pretend he agreed with it. He’d build
0:23:37 it in a different way. Then he’d make up an excuse why he did that, had to do it, you know?
0:23:41 And there’s all these disconnects where if you were able to use AI and be in total control
0:23:45 of a product, you will get closer to your actual vision for a product than dealing with
0:23:46 a whole group of people.
0:23:51 A hundred percent. I think the next like Stardew Valley or the next Minecraft is probably going
0:23:55 to come out of a single person just who’s passionate about making the game that they
0:23:58 envision and they’re just at home vibe coding it up.
0:23:59 Yeah, yeah, yeah, yeah.
0:24:05 But anyway, when it comes to the whole future tools website, I’ll just see this right now
0:24:09 in next week’s episode, I’ll break down my entire process to recode future tools.
0:24:12 So if you’re not subscribed, make sure you subscribe. I’m just saying that right now
0:24:17 while we’re waiting for our agent to go and finish up our stuff for us.
0:24:19 Waiting for it to do our work.
0:24:25 Yeah. So right now we’re on the third step out of about eight steps. It did its research
0:24:29 and now it’s gathering stock photos of shoes and potential models.
0:24:34 So it’s slow, but you know, these things will get faster and faster. Like probably they’ll
0:24:36 get like three to five times faster every year forever.
0:24:40 Yeah. Which means in like three years, we’ll have things like this where it’ll do like thorough
0:24:47 research and, you know, competitive analysis and check out your competitors designs and figure
0:24:50 out all that for you and put together and create a website in like a minute or something.
0:24:54 Yeah. Yeah. Once I see how it outputs it, I actually don’t know what the output’s going
0:24:59 to be because that’ll determine, you know, how you push it live. You know, I don’t know
0:25:03 if it’s going to give me like a zip file with all the files that created, or if it’s going
0:25:07 to actually like host it for me. Like I have no idea what the final output’s going to be.
0:25:12 I just know it’s not the fastest process in the world. I think once we get to the point where it’s
0:25:17 designing the HTML and CSS, it’ll speed up. I think the research is the slow part. I was just
0:25:20 looking at the terminal and it was actually like downloading images. I don’t know where it was
0:25:24 saving them to, I guess like the virtual computer. So it’s just copying somebody’s shoes.
0:25:29 Yeah. Well, I told it to use stock photos. So it just downloaded a bunch of stock photos.
0:25:33 It’ll probably have watermarks on them or something. Yeah. I can see now it just generated a bunch of
0:25:38 fake testimonials from Sarah Johnson, Michael Chen, Jessica Rodriguez, and David Wilson.
0:25:44 But now we’re on step four of eight. I do think once it gets to the HTML and CSS, that should speed up
0:25:49 because I know Claude is pretty fast at that. And if it’s using Claude, then it should pick up the
0:25:52 pace. So I guess we need to figure out what’s the key business takeaway for people.
0:25:58 In my mind, the key business takeaway for people is you could go and build the website,
0:26:02 that landing page, you know, that mini SaaS app that you’ve been wanting to build and you don’t
0:26:07 need to go hire a developer anymore. Right. Right. You can go and have a tool like Manus go and do the
0:26:13 research for you and build the bones for you. Like I’ve had software ideas over the years. I’ve been like
0:26:19 a solopreneur for the last 16 years now, not working for somebody else. And over the years, I’ve tried to
0:26:24 develop various softwares and I tried to get into the SaaS business and it was always such a pain to
0:26:30 work with developers or to try to teach myself to code. And now I know I can just go to some of these
0:26:37 AI tools and just say, here’s what I envision, go build it for me. Like, I feel like that’s valuable
0:26:42 to any business, especially the sort of, you know, small businesses and solopreneur types, which I think
0:26:44 is probably a big chunk of the people that listen to the show.
0:26:48 Yeah. You know, whether you’re in a big company or you’re solo, like, I mean, if you’re in a big
0:26:52 company, you may have a situation where you have like all these ideas, but there’s no way to express
0:26:56 those ideas or test those ideas or make them material in any way. Now with stuff like this,
0:27:00 you know, you could be like, okay, well you can’t do that during normal working hours, but at night
0:27:05 you could just like chat with AI and build like a prototype. We’re going to get closer and closer to
0:27:08 like where people have an idea and that the idea becomes real.
0:27:13 And then you can test to see if that idea was actually good or not really, really fast.
0:27:16 Where in the past, like if you had an idea, it’s like, well, okay, am I going to go spend like
0:27:21 months working on that idea? And then maybe nobody likes it. We’re going to get where you can just
0:27:24 have an idea, test it in a day, hours.
0:27:28 Have you ever read any of the Gino Wickman books, like get a grip or rocket fuel or
0:27:33 I haven’t, I haven’t traction. So he’s got this concept that every business kind of needs
0:27:38 two people to be successful. You need a visionary and you need an integrator. The visionary is the
0:27:44 person that has the big ideas that sort of sees where the company is going, pretty much has that
0:27:48 roadmap and vision of where they want to get it, but they’re not necessarily the person who’s going
0:27:53 to get into the weeds and write the code and build the website and, you know, do all the outreach.
0:27:58 They’re not the like doer in the company. So the company needs an integrator also. And an integrator
0:28:03 is that person that’s going to either go in, you know, hire and find the people that you outsource
0:28:08 to, or they’re going to be the person that develops the thing themselves, or, you know, they’re the ones
0:28:14 that gets into the weeds and sort of does the building. And to me, this really opens up the
0:28:18 possibility for like everybody to be that visionary, right? It really opens up the possibility to like,
0:28:25 I have this idea of where I want to go and I don’t need an integrator anymore. AI can be my
0:28:29 integrator, which is good and bad because everyone’s got a bunch of ideas and they’re not all good.
0:28:33 Yeah. And it’s also not great if you’re not a visionary type person, right? If you’re not an
0:28:39 idea person, right? That’s where I think there might be some struggle in the future with this AI era that
0:28:45 we’re entering into. I think this AI era that we’re entering is really, really, really good for the
0:28:51 visionary type people, but it could be really, really sort of catastrophic for the integrator type
0:28:56 person because the integrator type person are the ones that probably have to worry about, you know, what am I
0:29:01 going to do like with my skillset that I’ve built now that AI can do it. But if you’re that visionary, you’re
0:29:06 that idea person, you’re that thinker, the era that we’re entering with all of these AI tools, like best time
0:29:08 ever, best time ever for that person.
0:29:14 Yes. Yeah. Seriously. Like, like, I’ve dreamed of this my entire life. I’ve always had all these great
0:29:18 ideas. Sometimes the timing was wrong, you know, and I’ve always been like lightly technical and had to
0:29:23 deal with tons of engineers and designers and then raising capital for that. And just to think that now
0:29:30 any idea you have, you know, I kind of wish I was like, you know, 18 again, like right now, like really
0:29:36 wish I was 18. Because like, the fact that now or very soon, any idea you have, you can just create it and have
0:29:40 some level of control over the whole process. Like I said, like, you know, there’s no disconnect there. Like you
0:29:44 got an idea and then you deal with a designer and engineer. No, you can be part of that entire process to make
0:29:49 sure you actually create the thing that you envisioned. Yeah, yeah. So exciting. I’m kind of glad all this stuff
0:29:54 started to happen after I already had a wife and kids, because otherwise, I’d probably just be obsessed and never
0:29:59 leave my house. You wouldn’t have the wife and kids. I’m super happy to have my wife and my son. So yeah, I
0:30:03 wouldn’t change anything like the whole eight, wish I was 18. No, I don’t actually wish.
0:30:11 So right now we’re on the sixth step. It’s doing the CSS. All right. So just move to the seventh step. Now. Now it’s
0:30:17 just testing. So we should have something to show for this in, you know, any minute now. It actually says deploy
0:30:23 landing page and share with users. So it may actually deploy it for us. I’m curious if it’s going to like give us a link
0:30:28 that’s actually usable, or if like, it’s going to give us code that we need to go and like deploy somewhere else.
0:30:32 Yeah. Next step is it goes and markets it for you, sends you some money.
0:30:41 Yeah. Yeah. Yeah. Yeah. I mean, quite honestly, I bet you can say, go and create some Google ads for me.
0:30:47 So one thing about Manus though, is it’s not going to log into sites on your behalf. So let’s say you
0:30:53 ask it to go check your email for you and like reply to the 10 most recent emails and save them as a draft,
0:30:57 right? You can tell it to do that. But when it gets to the spot where it needs to log into Google,
0:31:02 it’s going to say, I need to hand this off to you, right? So it’ll hand it off to you.
0:31:07 You then go and enter your login details and then hand it back. And then after you enter your login details,
0:31:13 it takes it over again and starts working again for you. So you can’t like plug in your username and password
0:31:16 and have it go and log in on your behalf.
0:31:22 All right. It looks like we’ve got something here. It’s still finishing the test and validate landing
0:31:27 page, but I’m actually seeing a landing page on the screen. It called the website step style,
0:31:29 which is actually not a bad name, honestly.
0:31:34 Okay. Yeah. All right. So we’re getting a preview of this site, but I don’t have like a,
0:31:40 an actual like a link or place to check it out myself. It looks decent just from the previews we’re
0:31:44 seeing. Yeah. I mean, it’s not amazing, but it’s pretty standard. I mean, it’s, you know,
0:31:51 it doesn’t have like, you know, parallaxing background images and animations when you hover
0:31:56 and stuff, but it’s like a pretty standard looking e-commerce site. But I mean, here’s the thing though.
0:32:02 It went and found all the images. It wrote all of the copy for us. It found all of the prices for us.
0:32:07 It did the research on the popular shoes for us. Like the design is, you know, it’s standard,
0:32:11 right? Yeah. But if you look at all the other stuff that came before it, that’s the more impressive
0:32:18 part to me. We know that Claude and V zero and replet and some of these tools can do, you know,
0:32:24 okay, pretty decent looking designs. It’s all the stuff before the design that I think is impressive
0:32:27 now. Yeah. We didn’t really provide it much either, right? Like if you were doing this actual,
0:32:31 you know, a business use case of like building landing pages, you’d probably be providing an asset,
0:32:36 giving a clear list of like, here’s the kind of landing pages I want to create, maybe giving
0:32:40 it some resources or like examples. If you did that, it probably would be able to make like
0:32:44 tons of landing pages for you. And they’d probably be good enough to test if I had to guess.
0:32:49 For sure. So that seems like that’s really going to change marketing, like being a marketing manager
0:32:53 or whatever, like you have an idea for like, you know, marketing campaigns or like landing pages
0:32:57 you want to have created. Like now that person really could just do the entire thing in theory,
0:33:01 right? Like they come up with the idea and then just have AI create all the landing pages.
0:33:05 Eventually have AI help create the ads. Maybe it can’t do that well right now,
0:33:06 but I’m sure that’ll be coming.
0:33:12 You can actually have it log into like your Google ads account for you and then it hands it off.
0:33:17 You plug in your username and password and give it back. And it can probably go and write the ads
0:33:22 for you. I don’t see why it wouldn’t do that right now. You probably wouldn’t want it to just deploy the
0:33:26 ads without you double checking them first, but it would probably go and like fill out all the
0:33:30 details for you. Yeah. That’s probably a good experiment for people in business to be running is like
0:33:34 doing it. Like I’m a very limited amount. Like it’s very small budgets, like a hundred dollars or
0:33:38 a thousand or whatever is small for you. You know, one thing that I think would be interesting to test
0:33:44 would be like, can it go into like Canva and create an ad for me? Right. Can it like actually
0:33:49 like manipulate and drag the images around and like actually design an ad for me? That’s like a banner
0:33:54 ad or something. Why don’t we try that? I bet it can’t though. If I had to guess. Yeah. Maybe in one of
0:33:59 our future episodes. Right. Right. One thing I’d love to see Manus do is have some sort of like
0:34:03 environment variables that you can plug in, right? Where you can actually give it usernames and
0:34:08 passwords. And ideally it’s in some sort of secure location and, you know, save to your local computer
0:34:13 and maybe not save to Manus’s servers. Right. I don’t know how that would work exactly, but it would
0:34:18 be nice if it can like sort of do some of the logging in on your behalf. So it looks like it’s finishing
0:34:23 up the last step now. I mean, this is huge progress. I mean, like think about baby AGI. It would come up
0:34:26 with that list of things to do and it would go off and do some research. It would do some basic
0:34:29 research. Wasn’t that great. And then it would probably go to the next step and then just get
0:34:33 stuck. Like, okay, let’s go start making the website. And it was just like, how do you do that?
0:34:35 Yeah. Yeah. Yeah. It would get stuck very quickly.
0:34:40 Okay. So it’s asking me a question now. So it says Manus is attempting to deploy the service
0:34:45 located at blah, blah, blah to the public network. Please confirm if you agree by clicking the button,
0:34:50 allow Manus to deploy the application to the public network. Um, sure. I don’t know where
0:34:55 the public network is located, but it’s nothing sensitive. So I’m not too worried about it.
0:35:01 It’s a fake shoe website and website clones are going to get really bad. Yeah. Yeah. I mean,
0:35:06 you look at like blue sky, right? And blue sky just kind of looks like a clone of Twitter already.
0:35:10 Yeah. Yeah. Definitely. You’re going to see a lot more like that. I think where people are like,
0:35:13 we’re launching a new social media and all it is, is a clone of something else,
0:35:16 but they change the color scheme. Yeah. And that’s where you learn, like, you know,
0:35:21 network effects really matter and distribution really matters in those things. That’s like
0:35:25 actually one of the core things to them is not even the technology. It’s the, the users and the
0:35:31 distribution. Yeah. All right. So now it’s editing the to-do file. All of the task progress are checked
0:35:36 off. It showed a URL here a second ago, but then it disappeared. All right. So to successfully
0:35:42 deploy the shoe website landing page, ready to share final results with the user. All right.
0:35:48 Share them. Let’s see. Aha. Okay. So here’s what it says. The landing page includes all the elements
0:35:54 you requested, social proof, research on best shoes, images of shoes, customer testimonials,
0:36:01 view all files in this session. Okay. So it gave me all of the files. So if we did want to deploy these,
0:36:06 what would be the best way to deploy them? Would you say like, so that other people can
0:36:11 actually access this? So here’s what I would probably do. I would probably load it into a
0:36:17 GitHub repository. Yeah. And then attach my GitHub repository to Vercel. Yeah. That’s what I was going
0:36:20 to say. Probably not the easiest way though. Like for like most people, there’s, there’s gotta be some
0:36:25 other way. I know when you use replets agent, it’s very easy, but I don’t know if you’re not using
0:36:29 their agent, how easy it is just to push something alive like this. Yeah. That might be a little too in
0:36:35 the weeds for this one, but I would imagine if you wanted to, you could tell Manus to push this to
0:36:41 GitHub for me and then set up a Vercel account, but you would have to keep on taking control every time
0:36:46 it was asking for login details, but it would walk you through it. At least it would be like, here’s
0:36:51 the next step. I need you to log in for me. You log in, you hand it back. And then here’s the next
0:36:55 step. I need you to log in for me. You log in and then hand it back. And then it would actually do
0:37:00 the process of setting up GitHub, setting up for cell and pushing it all. So it says Manus has
0:37:06 completed the current task and is ready for new requests or revisions. So it didn’t actually seem
0:37:13 to give me a live link where I can go and view it, but it did give me all of the files here. All right.
0:37:20 So links, no items, code files, images, documents, all. So we do have all of the files here with the
0:37:27 research notes, testimonials, stats, index.html style script to do. Oh, here it is. I’ve completed
0:37:34 your shoe website landing page open. Okay. So it actually put it online for us. It’s at this URL.
0:37:39 I don’t know if that’s like temporary or if it’s like fully deployed for us and we can just like
0:37:47 change the domain name, but here’s the website step style shop collection, explore categories. So if I
0:37:52 click explore categories, it sort of auto scrolls down the page for me, the numbers speak for
0:37:59 themselves. It’s got the, you know, global footwear market 495.7 billion. So it actually has some like
0:38:05 research stats on the page about the shoe market featured collection. This is based on our research,
0:38:11 right? These are trending shoes, the retro runner sneaker, Python print pumps, soft leather loafers,
0:38:16 performance runners. And it gave you all the research files too, right? It gave you like all the stuff
0:38:20 it collected. And so that’s crazy because you could do this same kind of thing I’ve been using for
0:38:26 projects where you could then feed that content into an LM and like kind of have a conversation about it
0:38:30 and like, okay, let’s talk about the research and then how do we want to change things based on that
0:38:34 research, right? Yeah. And then you could start iterating on it. Use this like the baseline for the
0:38:39 starting point. I mean, none of the links work. So it seems to be just like a single page without,
0:38:45 you know, any internal links yet. Yeah. But you’ve got the bones of a front end landing page.
0:38:51 We’ve got the science of comfort with some research text here. We’ve got some fake testimonials that
0:38:56 it made up for us and added stock images. I mean, it actually looks pretty clean. I mean,
0:39:00 it’s not like the best looking website ever, but it’s clean. You wouldn’t look at this and be like,
0:39:03 oh, this site looks like crap. It’d be a decent starting point.
0:39:06 Yeah. You take something like this into, you know, cursor or something like that, and then
0:39:10 have a conversation with it about what you wanted to change. And then, you know, probably after a day,
0:39:14 it’d be pretty good. Yeah. And I mean, Hey, if you wanted to create just like a landing page that
0:39:18 collects emails, you know, or is in the marketing world, they call it a squeeze page, right? If you
0:39:22 wanted to create a squeeze page, that’s like a single landing page that collects emails,
0:39:29 this would be a pretty easy way to do it. So yeah, that’s Manus. I mean, I would say after seeing it,
0:39:33 do you have any like, you know, additional thoughts, but I feel like we pretty much shared most of our
0:39:35 thoughts as it was processing.
0:39:40 Yeah. I mean, I guess the only thing is I find it super interesting. I think that probably though,
0:39:44 if I had to guess, everyone’s going to copy it, like how this is working. It’s not clear that
0:39:47 Manus is going to be the winner in this space. I’d be shocked if open the eye and Claude and others
0:39:52 don’t do the same. Cause I think agents is going to be a core part of their business.
0:39:57 Oh yeah. A hundred percent. Well, you know, one of the things I mentioned was talking about some
0:40:01 alternatives, right? Just because there is a 2 million person waiting list and they’re only letting a
0:40:06 handful of people in every day. Well, this week, the week we’re recording the week before you’re
0:40:11 actually listening to this open AI just released some new features inside their API. Here’s the
0:40:17 blog post from March 11th, new tools for building agents. So in their API, they actually gave people
0:40:22 access to sort of create these like agentic features in the apps that they’re building directly
0:40:28 through the API. So agents easily configurable LLMs with clear instructions and built-in tools,
0:40:33 handoffs, intelligently transfer control between agents, guardrails, configure safety checks,
0:40:37 tracing and observability. Yeah. One of the big parts of the release. Well, one of them was
0:40:42 they updated computer use. Apparently the computer use is doing way better on benchmarks. I haven’t had
0:40:46 a chance to test it yet. So that’s one. I remember it went up by like 15 or 20% on the benchmarks,
0:40:51 I think. Yeah. Another one is like, uh, the agents collaborating with one another. Yes. Like they
0:40:56 apparently really improved upon that. And that’s part of this release. Yeah. But most of it seems to be
0:41:01 like through the API, right? I don’t know if there’s like a front end UI where you can just go and play
0:41:07 with this. So, yeah, I mean, as far as like alternatives go, you do have the chat GPT operator.
0:41:14 The open AI operator agent is currently available on the pro plan for $200 a month. However, there are
0:41:19 plans to eventually make it accessible to users on the $20 a month plan. So I wouldn’t really say this
0:41:23 is a great alternative because you got to pay 200 bucks a month to get access to operator, but this
0:41:29 will do some similar stuff. And then there’s also this browser use tool, which is an open source
0:41:34 tool that you can use. It’s actually a Chrome extension that you can use. They do have a free
0:41:39 open source version. I’ve played with it. This was the tool that I used in a previous live stream
0:41:43 that I did where I asked it to go to Amazon for me and buy me some toilet paper. And it went and added
0:41:48 $600 worth of toilet paper to my cart. And then I canceled it before it actually placed the order.
0:41:53 So there is that you can go play with it, but browser use is actually what Manus is using under the hood.
0:41:59 I just think Manus is a lot better at sort of giving it good instructions where just giving it broad
0:42:04 instructions, like go order toilet paper is not clear enough, but Manus, I think using
0:42:09 the underlining like Claude or Quinn models is better at making sure it won’t go do stuff like
0:42:14 that. There’s also this other one I came across called highlight AI. I haven’t played with this
0:42:18 one yet myself, but they say everything you love about cursor across every app on your computer.
0:42:24 So if you look at the app, it’s basically a tab that opens up on your computer and you can ask it to do
0:42:29 stuff and it will open up different apps and run programs for you and stuff like that. Again,
0:42:33 not what I’ve tested, just wanted to show it else is available out there. And then there’s this one
0:42:41 called pig at pig.dev and it says operate windows apps with AI pig makes it easy to build complex
0:42:46 automations powered by AI. And this one actually will take control of your mouse for you. So it’s not
0:42:51 using like a virtual browser. So I don’t know how much I would trust this one yet. Cause it’s literally
0:42:57 taking control of your own computer, not a virtual computer. This one though, you do install and I
0:43:03 believe it’s open source. I’m not a hundred percent sure on it. I think they could have picked a better
0:43:07 name for something that you’re going to trust with your computer to like run your computer calling this
0:43:13 pig. Yeah. So it does say a pig lit is fully open source and you can check out the code on GitHub.
0:43:20 So pig lit is, I guess the open source version of this. Again, this one actually takes control of
0:43:29 your computer where computer use and operator and menace all seem to use like a virtual desktop that
0:43:35 it uses that it will go and take the actions inside of. I think for now, the virtual desktop is the
0:43:39 better way to go. Like when these things get really good and you can actually trust what it’s doing.
0:43:44 I think giving it access to your computer will probably make sense, but for now, I think I
0:43:49 would stick with the ones where the virtual sandbox. Yeah, I agree. I’m too scared to put pig on my
0:43:54 computer and a, it’s a tool called pig. Like you mentioned like that just doesn’t make me too
0:44:01 confident in it, but B I don’t want an AI going and controlling my mouse and give me back my mouse.
0:44:04 I can’t do that, Matt. You know, like I don’t want that scenario.
0:44:09 We’re locked in, Matt. You told us to do this, Matt. We’re going to finish it.
0:44:17 Exactly. But anyway, that’s a sort of breakdown of the current standing of the world of AI agents.
0:44:21 Obviously, Manus is the big talk of right now. A lot of people creating some really cool stuff.
0:44:25 I’m fairly impressed with it. I wish it was a little bit faster right now. It feels a little
0:44:30 bit slower than it should be, but that could just be a symptom of so many people all trying
0:44:33 to use it at once, right? If it wasn’t so bogged down, it might go a lot faster,
0:44:38 but it is a big moment because like agents are working now. Yeah. Like there’s been hyper agents
0:44:41 and it’s like everyone was saying like, this is going to be the year of agents. They are working
0:44:47 now and we’re only in March. Yeah. It is officially the year of AI agents. They are here. Yes. And we’ve
0:44:51 got like over nine months of more improvements. So I mean, like by the end of the year, agents are going
0:44:56 to be amazing. So anyone business, I think that’s what you should be thinking about is like be testing
0:45:00 these things out to like understand how they work and then try your best to extrapolate and like think
0:45:04 about over the next year. What does it mean when this thing is like five to 10 times better? That’s
0:45:08 most likely going to happen. And how does that impact your business? And how can you be using it?
0:45:14 Because that’s what I’m trying to think about daily. Absolutely. Absolutely. Well, I think that about
0:45:20 covers everything we can kind of possibly talk about around Manus and the current state of AI agents.
0:45:23 So we can go ahead and wrap this one up here. If you like episodes like this, make sure you’re
0:45:28 subscribed either on YouTube or wherever you listen to podcasts. We have a lot more stuff like this
0:45:32 coming next week. I’m going to share my whole breakdown of how I’ve been rebuilding future
0:45:36 tools. We’re going to talk about some other really cool apps that have been coded with AI recently and
0:45:41 break them down a little bit better. So that’ll be a fun episode that we’re releasing next week. So
0:45:45 again, don’t forget to subscribe and thank you so much for tuning in. We’ll see you in the next one.
0:45:47 Thank you.
0:46:03 I’ll see you in the next one.
Episode 50: What can you do with the new AI agent, Manus, to revolutionize your business? Matt Wolfe (https://x.com/mreflow) and Nathan Lands (https://x.com/NathanLands) are diving into the AI sensation Manus AI.
This episode breaks down the creation and capabilities of Manus AI, including how it can be used to build a functional website seamlessly. Explore the impressive tools and unique workflow of Manus, which brings AI agents to the forefront by autonomously handling complex tasks. Delve into the implications for visionaries and integrators, and see how new alternatives are shaping the AI landscape, promising more productivity for businesses, solopreneurs, and beyond.
Check out The Next Wave YouTube Channel if you want to see Matt and Nathan on screen: https://lnk.to/thenextwavepd
—
Show Notes:
- (00:00) Server Overload Halts AI Demo
- (04:58) Effective AI Navigation Test
- (09:21) Project Astra Glasses as Teleprompters
- (11:20) Isolated Sandbox Sessions Explained
- (15:21) Focused AI Specialization Triumphs
- (18:51) Website Editing Transformed by AI
- (20:08) Decentralized Game Design Evolution
- (24:37) AI Tools for Easy App Building
- (26:13) Visionary and Integrator Dynamics
- (31:08) Automating Landing Pages in Marketing
- (35:11) GitHub and Vercel Setup Process
- (38:36) OpenAI API Unveils Agentic Features
- (40:37) AI Tools for Improved Task Management
- (43:54) AI Discussion Wrap-Up & Upcoming Preview
—
Mentions:
- Manus: https://manus.im/
- Highlight AI: https://highlightai.com/
- Claude: https://claude.ai/
- Quin: https://www.quinengine.com/
Get the guide to build your own Custom GPT: https://clickhubspot.com/tnw
—
Check Out Matt’s Stuff:
• Future Tools – https://futuretools.beehiiv.com/
• Blog – https://www.mattwolfe.com/
• YouTube- https://www.youtube.com/@mreflow
—
Check Out Nathan’s Stuff:
- Newsletter: https://news.lore.com/
- Blog – https://lore.com/
The Next Wave is a HubSpot Original Podcast // Brought to you by The HubSpot Podcast Network // Production by Darren Clarke // Editing by Ezra Bakker Trupiano