I Built an AI Agent Company (From Scratch)

I sit down with Dotta, the pseudonymous co-founder of Paperclip, the open-source agent orchestrator that exploded to 30,000 GitHub stars in under three weeks. We walk through a live demo where I pick a startup idea from my idea browser and we spin up a full AI-agent company in real time — hiring a CEO, founding engineer, QA agent, video editor, and content strategist inside Paperclip. Dotta shares practical tips on agent configuration, memory systems, skill installation, and the "Memento Man" mental model for keeping agents on track. The conversation covers everything from token spend management and agentic design patterns to the future of importable, shareable companies and the upcoming Maximizer Mode. Skills to build your agent team: https://startup-ideas-pod.link/skill-suite Timestamps: 00:00 Intro 02:32 What is Paperclip 04:21 Choosing a Startup Idea for the Demo 05:48 Setting Up your agents 07:51 Hiring Your First Agent and Creating a Plan 12:39 Agent Configuration and Persona Setup 17:08 Skills: Installing and Managing Agent Capabilities 21:02 How to Get Top-Quality Output from Agents 24:05 Token Spend Tracking and Subscription Usage 25:49 Agentic Design Patterns and QA Loops 29:05 Taste and Values: What AI Still Cannot Do 30:09 How Many Agents Run the Paperclip Project 32:32 Routines: Automating Recurring Agent Tasks 36:36 Who Is Using Paperclip Today 38:57 Shareable and Importable Companies 42:49 The Unproven Frontier: Do Agent Orgs Actually Work? 42:49 Maximizer Mode and What's Next 44:29 Did Dotta Expect It to Go This Viral? Key Points Paperclip is a bring-your-own-bot orchestrator: it works with Claude Code, Codex, OpenCode, and any model on OpenRouter, so you are not locked into a single provider. AI agents are "Memento Man" — they wake up capable but with zero memory, so you need heartbeat checklists, persona prompts, and written context to keep them effective. The biggest lever for quality output is encoding your own taste and values into agent skills and brand guides, because AI can do everything except know what you actually want. Agentic design patterns like engineer-to-QA review loops matter more than one-shotting an entire startup; structure prevents compounding errors. Paperclip tracks every token spent and every task completed, solving the problem of running dozens of agent windows with zero accountability. Importable, shareable company templates (like Gary Tan's G-Stack or a full game studio) point toward a future where you "aqua-hire" proven agent teams instead of building from scratch. The #1 tool to find startup ideas/trends - https://www.ideabrowser.com LCA helps Fortune 500s and fast-growing startups build their future - from Warner Music to Fortnite to Dropbox. We turn 'what if' into reality with AI, apps, and next-gen products https://latecheckout.agency/ The Vibe Marketer - Resources for people into vibe marketing/marketing with AI: https://www.thevibemarketer.com/ FIND ME ON SOCIAL X/Twitter: https://twitter.com/gregisenberg Instagram: https://instagram.com/gregisenberg/ LinkedIn: https://www.linkedin.com/in/gisenberg/ FIND DOTTA ON SOCIAL X/Twitter: https://x.com/dotta Paperclip: https://paperclip.ing Github: https://github.com/cryppadotta

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Greg Isenberg?

  How do you turn a group of AI agents into a company that actually runs itself? You know, with org charts, roles, goals, budgets, and agents working around the clock 24/7? There's this project called Paperclip. Maybe you've seen it because it just blew up. It got 30,000 GitHub stars in the last three weeks. And it's an open source project that's building the orchestration layer for exactly that, hiring a team of AI agents. So I sat down with Doda. He's the creator. And it's one of the first times he's publicly talked about how to use his product. We broke down exactly how to use Paperclip in practice and what a zero-human company could actually look like using Paperclip. Enjoy the episode.

• 0:53Greg Isenberg?

  Doda is here. Well, his AI avatar is here. And we're going to be going through Paperclip, the viral, you know? What are you calling it?

• 1:02Speaker 01

  Yeah, so it's an agent orchestrator for zero-human companies. So it's a tool that you can use to put in your ideas and manage a company of agents.

• 1:15Greg Isenberg?

  OK, and by the end of this episode, what are people going to get out of this?

• 1:19Speaker 01

  Yeah, so by the end of this episode, you will be able to know how to get Paperclip up and running and use it to manage a team of AI agents to help you with your business.

• 1:30Greg Isenberg?

  OK, and you're not going to hold back from people, right? You're going to, you're going to, you're going to give us the unfiltered how to get started, how to get the most out of it, and maybe where the future of this product is going, right?

• 1:41Speaker 01

  Absolutely, yeah. I'll give you my best tips. You know, I think there's a lot of tricky things when it comes to using AI agents, and we'll talk about it. We'll talk about ways to keep them moving. We'll talk about ways to get quality work. I'll give you all my best details.

• 1:55Greg Isenberg?

  And, and just, just, I have to ask, like, why are you an avatar? And why are you not showing your face?

• 2:02Speaker 01

  Oh, yeah. So before I worked on Paperclip, I was working in NFTs. And the Dota persona has just been my online persona ever since.

• 2:14Greg Isenberg?

  OK.

• 2:15Speaker 01

  You know, Naval always says, "The best, the best outcome is to be famous, but nobody knows you."

• 2:24Greg Isenberg?

  I understand. And listen, you got good hair. So, you know, I'm not, I'm not judging. All right, let's get into it.

• 2:32Speaker 01

  All right. Yeah. So let me talk about Paperclip. So Paperclip, here, I can show you this dashboard. This would be the dashboard that I use to manage Paperclip itself, which is pretty involved. So let's take a step back. So the idea with Paperclip is it is a tool for, the tagline is zero human companies. Now, I think that is maybe a bit aspirational. I'm sure you guys have seen tools like Pulsia, where you, you, you give it your credit card, it creates business ideas for you. And Paperclip is designed to be have a little bit more control right now because it's designed for

• 3:10Speaker 01

  work that you're really accountable for doing. The idea behind Paperclip is that you're really managing your business goals. So it's not, you know, on one end, you might have something like Pulsia, it's totally automatic. On the other hand, you might have like an AI coding tool where there's a bunch of tabs open and you're managing pull requests. With Paperclip, your idea is you're going to manage business goals. You define your goals, you hire a team, and then you approve what they're doing and they go and they work on it. So an interesting thing about Paperclip is that it's bring your own bot, right? So you're already starting to see the labs come up with project management tools. You have OpenAI, you have Anthropic. With OpenAI, they have Symphony. With Anthropic, of course, they're releasing new project management tools every day.

• 3:54Speaker 01

  With, with Paperclip, you're able to bring your own bot. If you use Cursor Cloud, if you use OpenClaw, if you use Claude CLI. Like, I would say that all the programmers that I know, or all the like entrepreneurs that I know, are using lots of different models because they all have such a different personality. And that's one of the key things about Paperclip is that you can bring your own agent into the system. So, yeah, Greg, one of the things I thought we could do was we could go to your idea browser.

• 4:28Greg Isenberg?

  Yep.

• 4:28Speaker 01

  And we could look here and we could just try to decide something from the idea browser that we could try to use as a case study. So I don't know, what's, what's one of these that you think might be a good, a good option?

• 4:48Greg Isenberg?

  I mean, there's a few really interesting ones like I like the you tell me I'll give you two and you pick so scroll up.

• 4:53Speaker 01

  Yep.

• 4:54Greg Isenberg?

  So if you go down, there was a finance finance app that builds money habits in three minutes a day.

• 5:02Speaker 01

  Okay.

• 5:02Greg Isenberg?

  That's one option. And then the other option, if you scroll down.

• 5:06Speaker 01

  Yep.

• 5:08Greg Isenberg?

  The flight compensation recovery agent for frequent flyers.

• 5:13Speaker 01

  Okay.

• 5:14Greg Isenberg?

  Are great ideas.

• 5:16Speaker 01

  Okay, nice. Yeah, let's try. Let's try a finance app that builds money habits. So one of the things I noticed actually when I was testing this out, if you just right click, you can copy like pretty much a good description from this more than more than this. So here we can come in here like, what do we want to name our app? We'll name it.

• 5:35Greg Isenberg?

  Moolah.

• 5:36Speaker 01

  Moolah. Okay, great. And what is this company trying to achieve? Right? So this text box, I suppose is a little small, but we'll just paste it and we'll come back and we'll look at this in a second. After you write down what you want your company to do, you create your first agent. So Paperclip right now best works when it's on your local machine, and especially if you already have something like Claude Code or Codex installed. I think these tools are really important for really any agent driven entrepreneur. I'm guessing pretty much everyone in your audience has at least like Claude Code or Codex installed, right? We recommend that you use one of these two for your CEO, which would be your first agent, but we do support other agent types, right? You can use Open Code, which means

• 6:19Speaker 01

  you could use really any model that you have on Open Router. So you've probably seen Open Router. They have like, what is it leaderboard that kind of shows the most popular? Yeah, the rankings. They have the AI model rankings. You can use any of these models with Paperclip. I think. You're going to get the best results with Paperclip if you use a frontier model for your CEO, but then maybe for other tasks, you can get away with a cheaper model. A hot tip is OpenRouter actually often has free models. So there's been, this Hunter Alpha model was free for a couple of weeks, the StepFlash free. So if you have work that can be accomplished by a free model, right?

• 7:02Speaker 01

  Which would be sort of, you know, not all work can, you can use it and get free tokens. Also the Pi coding engine, whatever. So it basically depends on some of these existing agents to operate. This one's actually going to run on my local machine, so we'll just go next.

• 7:20Greg Isenberg?

  And do you recommend, you know, doing local? Or how should people think about this?

• 7:27Speaker 01

  Yeah, so right now, Paperclip mostly works when you run it locally. We're working on doing a cloud deployment. So either you can self-host it yourself, because it is open source, or eventually we'll provide a hosted solution for people that want to do that. But I think local is really how most people are doing a lot. Everything with Paperclip right now, yeah.

• 7:50Greg Isenberg?

  Cool.

• 7:51Speaker 01

  So the first thing we want to do is give our agents something to do. Hire your first engineer and create a hiring plan. So you are the CEO, you set the direction for the company, hire a founding engineer, write a hiring plan, break the roadmap into concrete tasks and start delegating work. So you can edit this if you want. You know, maybe your business doesn't need an engineer. Maybe you're like doing marketing, sales. You can sort of adjust this depending on what you want. But, you know, having a founding engineer is helpful because almost every sort of online business needs to do a bit of coding. So we'll click next here. Okay, we've got our company, our plan, and it kicks off. So what you have here is along this sidebar, we've got different companies.

• 8:34Speaker 01

  Paperclip is made to run as many companies as you want to. So I'm running Paperclip itself in here. I'm running Forgotten Runes from here. I've got a couple other companies that I'm like, you know, different apps that I'm working on.

• 8:50Speaker 01

  And here we've got our new entry for the moolah company. You can see right here, we've got this dashboard. Part of the thing with Paperclip is that it will track the monthly spend and it tracks all of the work your agents do. I mean, I built Paperclip because I was using Claude Code to build my companies and I would have 20 or 30 Claude Code windows open all at once. And I couldn't remember what any of them were working on. I would set them to run over the weekend and I'd come back and I had no idea what anyone did. I couldn't remember what they were supposed to do and they had spent all my money on all their tokens. My budget was busted and I didn't know what was actually accomplished. And so Paperclip is designed to solve those problems.

• 9:35Speaker 01

  It actually tracks your monthly spend that you have in tokens. Everything that happens is done through issues. So right now, we have this first project here and it hires our first engineer. You know, these things sort of take a while. Here we go, we can see we've already got our founding, our hiring plan installed. So immediate hire, founding engineer, you know, a roadmap, breakdown, et cetera, et cetera. You know, what it actually is for this call doesn't really matter, but you could look in here, you could edit it, give it feedback, give it comments. So what happens in Paperclip is your agents

• 10:18Speaker 01

  have tasks that get created and then they work on them. I mean, that's, it's as simple as that, right? And Paperclip sort of orchestrates this idea that each agent, each task can only be handled by one agent at a time. So you're not going to get this kind of conflict of agents stepping on top of each other. Okay, look here, we've got something in our inbox. Okay, hire agent, our first engineer. There's a request from our CEO to hire an engineer. Now, as you get more comfortable with Paperclip and how your agents are acting, you can turn this off and let your agents hire agents automatically. But for sort of these like defaults, we've said, you know, a lot of people are getting started their first time.

• 11:02Speaker 01

  We want to approve this manually. So we just approved that our CEO can hire his first engineer.

• 11:11Greg Isenberg?

  So is that sort of the big mindset shift with Paperclip? Instead of, you know, thinking of jobs to be done, you think of hiring agents?

• 11:23Speaker 01

  Yeah, I think one of the things, instead of thinking about like the details, the goal of Paperclip is that you're thinking at a higher level. Like you think of yourself as like the board and you should be talking to the CEO and saying, here are high level goals, here's the MRR that we want to reach, here's like what we want to achieve. Now I'm asking you to make a plan and institute what needs to happen within the company.

• 11:54Speaker 01

  So you can see here, we've hired this engineer and now we've got like five live tasks actually happening here. So let's look in a bit more deeply and see what's going on. So if we click on the issues, you can see we've got more issues. Set up the scaffolding, continuous integration, do progress tracking, the user auth and onboarding, build a core daily loop, right? So the agent right now is working on these things, the engineer is. Now I can see right now that the engineer is actually just working on one of these. So I might go in here and I can up the concurrency, so whatever. There's advanced settings in here where you could say, you know, run four engineers at once and it's a bit in the weeds, but.

• 12:39Greg Isenberg?

  So wait, going back to that, because I know it's the weeds, but the configuration, like should we spend any time in here?

• 12:47Speaker 01

  Sure.

• 12:49Greg Isenberg?

  How should we be thinking about this?

• 12:50Speaker 01

  Yeah, so one of the key things that you want to be able to do with your agents is you need to configure them with a bit of persona. So the Paperclip right now gives you a default persona for your CEO. So.

• 13:09Speaker 01

  If you look here, it says, you know, you are the CEO. Here's how we expect you to use memory. We use a memory system by default that just stores your memory in files. That was designed, it's using like the Para Memory System, which is like, I think, Diego Forte's invention that this skill is from Nat Eleyson. He runs the Felix Bot. So it gives your CEO a default memory. Now, Agent Memory is a really involved topic. There are a lot of different choices, which I'm happy to talk about. But this is gives you a simple default. And then you give it some safety considerations, right? Don't exfiltrate my secrets. Here are some other things that I want you to look at.

• 13:54Speaker 01

  One of the key things that you want to look at is the heartbeat. And here's the CEO's heartbeat checklist. So when you think about AI agents, here's a tip about using AI agents with Paperclip or anything. Have you ever seen the movie Memento? Of course. Yeah, yeah. So if for people haven't, Mento, the Momentum Man, he wakes up every morning and he is like really capable, but he can't remember where he's at or where he's doing. And he has to leave himself a little notes, right? Your AI agents are Momentum Man. They wake up, they know how to fight, they know how to drive, they know how to like, just take care of themselves and spend money, but they don't know who they are. They don't know where they are. They don't know what they're supposed to be doing.

• 14:38Speaker 01

  And so what you need to do is actually write down, you know, give them little Polaroids or write tattoos on their arm on what, who you are and what you're supposed to be doing. And so the heartbeat basically means like, when you wake up as an agent, what are you supposed to do? And, and this gives you these instructions. So, so fetch Agents Me, confirm that you, who you are, learn who you are. Read today's plan. Look for your assignments. Check out the work and then break it down into tasks and extract your memory and then you're done, right? So these are your responsibilities. I would say that, you know, Paperclip has been out about three weeks and we are already

• 15:23Speaker 01

  improving on this on a daily basis. You know, one of the key things that you always have to do with your

• 15:32Speaker 01

  Agents is you really want to set up a way to kind of evaluate the quality of the prompts that you're giving them. And this is really like a new field that a lot of people haven't, it hasn't been figured out yet. We'll be building more tools in Paperclip for helping you do that. But really, how a lot of people just do it practically is when it does something you don't like, you come in here and you say, "Rule," like, like, "Make sure you remember to set," you know, like, "If it's forgetting to set like a win condition, like, make sure you set like a success condition for every task and ask the developer to have QA review it," right?

• 16:13Speaker 01

  Like, "I might, I might be using this and I realize he never defines what success looks like and he never asks QA to review it. And I'll literally just like add that line in here and then save it. And then, you know, typically you'll, you'll get a bit better results." Which, by the way, I think is something that we should have our CEO do. So I'm going to ask him to hire a QA person to help review the work. So that's another thing, like, so this is another way that you use Paperclip is whenever you have an idea for something, just ask. The CEO knows how to use Paperclip very, very well. And i- if you're having trouble, if you want to set up a

• 16:55Speaker 01

  new project, you want to set up, you improve your organization, you want to make a plan, you don't, you, you need advice, you just ask your CEO to do it. So I'll create this, this issue, let him work on that. Another key way that you might want to configure your agent would be with skills. So, you know, if you've tried to use, you know, OpenClaw, for example,

• 17:22Speaker 01

  Okay, so let me talk about OpenClaw for a minute. OpenClaw is insanely powerful, right? It's such a beautiful piece of software, just taking the internet by storm. It's really the first time that, you know, if ChatGPT was the first time people realized that AI could talk, OpenClaw was the first time people realized AI could act, like that you could actually have AI act on your behalf. I would say that like Paperclip is aiming to be the third

• 17:56Speaker 01

  moment where you realize AI can do real work that you're accountable for, but that's the future. Like with OpenClaw, one of the things that's amazing about it is that it can just do so much for you automatically. But the problem that people run into is like, if you use OpenClaw for two or three or four weeks, you start to realize that it kind of like falls apart. You don't really know what it's doing, you don't really know what it can do. It breaks and you don't know how to fix it. And so Paperclip can use OpenClaw as an agent, but Paperclip is designed to give you a bit more accountability into what everyone's doing. You can break your agents up into different personas and you can also give them different skills.

• 18:40Speaker 01

  The best way to find skills would be skills.sh, maybe not the best way, but it's one of the most popular. Have you used this before?

• 18:49Greg Isenberg?

  Yeah, my question here is, how do you know you're not installing skills that are, you know, have bad stuff in it?

• 18:58Speaker 01

  It's a problem.

• 18:58Greg Isenberg?

  Malicious stuff.

• 18:59Speaker 01

  Yeah, it's a real problem and it's something you have to be careful with. I don't think anyone solved that.

• 19:06Speaker 01

  One of the things that skills.sh has is they do have some degree of security audits.

• 19:16Greg Isenberg?

  Yep.

• 19:17Speaker 01

  So they have these badges. So I think that that's as best as you can do probably for now.

• 19:24Greg Isenberg?

  And I mean, you can, just because it has a lot of GitHub stars doesn't mean it's, you know, 100% secure, but it does give you directional kind of, it's a data point, right? If it has, if the re-motion, you know, skill has 100 plus thousand stars, for example.

• 19:45Greg Isenberg?

  Chances are, it's likely that there's value there.

• 19:49Speaker 01

  That's right. That's right. So what we can do if we wanted to, if we did want to make a video about this, right, we can copy this here and we can go to Paperclip. There's a couple of things that you could do. You can, your skills are organized by companies. So we should be able to paste this here. Or if you can't remember that, you can just ask your CEO, like, add the re-motion skill or whatever. But, you know, we don't need to do that right here. We can just click add. Yeah, there we go. So we've, now we've got the re-motion skill installed into our company. Okay? And this actually keeps track of, like, you can check for updates if the skill gets updated, et cetera, et cetera.

• 20:34Speaker 01

  You know, something that I might often do, like I might do is I might say, you know, hire a video editor and give them the re-motion skill and I would paste the URL. And the CEO knows, if you sign this to the CEO, he knows how to do this.

• 20:52Speaker 01

  For the sake of demo, we'll just wait because we don't need to do that this second. But I'm just telling you that's, that's, that's how you use Paperclip. You just ask the CEO to do things for you.

• 21:02Greg Isenberg?

  So I have a question, I have a question about that specifically. So, you know, the re-motion skill, you know, I've used it and I've gotten, sometimes I've gotten good results and sometimes I've gotten bad results. And the way I've gotten good results, the best possible results, is giving it the most possible context, the most possible references. And so my question is, you know, if I'm going to hire a video editor and give them the re-motion skill, how do I ensure that this is the, you know, a top 1% video editor using Paperclip?

• 21:35Speaker 01

  So you, you can never get around the idea that you have to provide that context. Ultimately, like, Paperclip is not going to, on its own, write that context for you. That said, it does have tools that can help. So when you have issues, you can have them, we can hire our QA agent.

• 22:05Speaker 01

  In your issues, we do have documents. And we're going to add features to do a better job at like organizing the documents, organizing the artifacts that you have within Paperclip. We'll do that. But of course, your agents are able to access pages anywhere you want. So for example, on the Paperclip website, I have a brand guide. Yeah, the brand identity. So when we're asking this question, which is like, how do I make sure that my agent makes a video that's good? One piece of that is like, do you have a brand guide yet? And you can ask your agent to do this, right? So I would say for kind of Moolah, we haven't necessarily gotten there yet, right?

• 22:50Speaker 01

  Our agents are still kind of working. But eventually, what you would do is you would create an issue, create me a brand guide, and it would give you the URL and you would look at it. Then the other part would be, as I might say, well, why don't we do this? Right? Let's say, let's plan a 60-second remotion video about Moolah showing the key pitch that we could share on Twitter. And we'll assign this to the CEO for now, because we don't have a CMO, but maybe we should. Maybe it should be in another project, but whatever, we'll get there.

• 23:32Speaker 01

  So I would say, make a plan for this. Don't make it yet, but we need a video demoing the product and showing the pitch. It needs to be animated. Needs to be animated and pretty. Agents don't really like care about typos, so you don't have to care that much about it.

• 24:00Speaker 01

  Whatever. We'll let that, we'll let that run and we'll come back a bit later. One of the things you'll notice is that our monthly spend right now is $0, even though we've been doing all this work. Part of that is because I'm using my Codex subscription or Claude code subscription. Actually, I'm not even sure which model we're using for these. It might be Claude.

• 24:21Speaker 01

  Oh yeah, look at this. Our agent, yeah, so our, our, our, our CEO is using the Claude Opus and it's, you are able to use your subscription as well as API credits if you want. And that's why you're seeing here that the costs are zero. It's a little bit of a, you know, it's, you're still using your kind of inference for your subscription. But if you hooked it into something like Open Code or we're using like pure pay per token values, you'd get kind of real dollars here.

• 24:55Speaker 01

  Okay, we've hired our video editor and let's check and see if he was given the skills. Yeah, so there you go. He has the ReMotion Best Practices skills enabled. Our CEO is still working on planning the video, so we'll just let him kind of, kind of work.

• 25:18Greg Isenberg?

  Doesn't seem like you have the context of firing a CEO, firing an engineer, firing a QA agent. Is that correct?

• 25:25Speaker 01

  No, so yeah, you, you can. Yeah, you can come in here to configuration and let's see. Oh, where's, oh, let's see here.

• 25:33Greg Isenberg?

  I mean, I hate to be that guy, but.

• 25:35Speaker 01

  No, right here, we got terminate. You can terminate. Yeah.

• 25:38Greg Isenberg?

  Okay, well, don't, don't do it, but you know.

• 25:42Speaker 01

  He hasn't had a chance. Yeah.

• 25:43Greg Isenberg?

  Yeah, give him a shot. Give him, give him a 30 day cure period at least.

• 25:48Speaker 01

  Right. Yeah. So, um, some of the other things that you can do with Paperclip is this idea of shareable company. So let me take a step back, which would be, it is not that clear how to structure your company in a way that operates really cleanly and creates good results. Right? Everybody who's sort of tried to one-shot a new startup with AI, you realize like it's super fun for the first half an hour and then it just kind of falls apart. There are a lot of new patterns that are coming out around like agentic design patterns. OpenAI has a blog post where they call it harness engineering,

• 26:33Speaker 01

  but it's just a fancy way of saying like ways to have your agents interact with one another that will help you get better results than just trying to quote, quote, "one-shot" results. So, these patterns are still being developed. I would admit that I'm even learning them as I go. A really basic one would just be like having a process within your organization where after the engineer creates something, he asks QA to QA it. This is especially important for like web apps where you need to like click on something and make sure that it actually works. So for example, you know, in here, you can like @mention the CEO or @mention a project, you know, like the onboarding project. And this

• 27:18Speaker 01

  was like a really, you know, you see it all the time, but for whatever reason, my agents had a hard time coding this. Funny, I can even see here that the that the pills have a different height and that really annoys me. So like what I might do is I would go into Paperclip and I would actually post this screenshot here and I would say, "The agent mentions pill is misaligned with the like project mention pill." Fix it, right? And I'm going to assign this to the Claude coder for the Paperclip app

• 28:01Speaker 01

  and I'll say, you know, "When you're done, pass it to QA to get eyes on it and verify it works here with a screenshot." Now, listen, if you're going to run a zero-human company, you cannot be managing your apps at that level, right? Like, I'm actually acting not even as a board member or CEO, I'm actually acting as like, like a design PM or something. And so, there is kind of this like structure that needs to be built where you should have a design agent who's already looking at the website and detecting that stuff. And, and so that way, you don't have to be kind of...

• 28:49Speaker 01

  Micromanaging your agents. And I would say we'll get to that, right? You basically have to make sure that you're ready to spend the tokens to do that. And you also need the evals to make sure that your designer sort of shares your values in terms of how you want the app to work. I would say that like even the models today, the best like GPT 5.4 or like OPUS 4.6, some of their taste is still not quite there. And, and that's why, that's where a lot of the secret sauce is. That's where you're actually going to write your own skills. And that's where, because you're basically going to impart your own taste. Like AI can do everything except know your values.

• 29:32Speaker 01

  And so you actually have to become more aware of your values and find out how to communicate them back.

• 29:42Greg Isenberg?

  Which is even in a pre-AI era, the, the concept of a good leader, of a good manager, of a good CEO, of a good founder, is very much someone who can clearly communicate their values and taste, right?

• 29:59Speaker 01

  That's right. That's right.

• 30:01Greg Isenberg?

  So not much has changed, except the vehicle to, to doing it has changed. Instead of hiring employees, you're hiring agents. I had a question before, before we move on. Now I had a question, if, if you, you know, or if you go back to the paperclip project, which by the way, it's so cool that you use paperclip to build paperclip. I expect nothing less from you, obviously. But, you know, you were talking about, you know, it's important to hire, you know, an agent for this and an agent for that. How many agents do you have for the paperclip project? Like?

• 30:35Speaker 01

  Yeah. Not that many, actually. Um, so this would be the org. It's actually a little hard to see here. Um, and yeah, I would say that it is growing all the time. Um, so we've got the CEO, reporting to him would be a CMO, a UX designer and a CTO.

• 30:57Speaker 01

  And I use different agents for different reasons. I would say that the Cursor Coder and the Claude Coder would be my kind of workhorse. They're using the Frontier models on my subscriptions. Then I have a QA engineer who has both the Claude browser installed, as well as there is a skill called Agent Browser, which is, it basically gives your agent access to a web browser in a way that is a little bit faster than using Chrome. Like, if you've used Chrome with Claude code before, I'm sure you've seen it, like, pops up a Chrome window and like takes over your computer. And if you click on it, you mess it up. And so that's the QA.

• 31:42Speaker 01

  I also have, I'm starting an evals engineer. So evals would be basically just a way of saying like, how do we start to kind of, like, how do we do performance reviews on our agents? And be able to go back and reflect on what they did. Paperclip is going to be building tools where you can look at your past issues, look at the feedback that you've given your agents, and your agents will be able to learn from that, right? If you find that you're giving the same feedback over and over, we're going to learn that and you won't have to make the same mistake twice. We've got a UX designer who kind of has context around what I like for design. I would say that's definitely a work in progress. And then we're starting to build out the marketing organization, right?

• 32:26Speaker 01

  Starting to build a blogger, a content strategist.

• 32:31Speaker 01

  You know, one of the things that we just added this week is the idea of routines. So I would say that like, using Paperclip to manage Paperclip is not where I want it to be. For example, we're adding new things to the code every day. We need to have someone who looks at all of the things that changed in code and crafts a Discord message that we can share in the Discord. In fact, we should hook it up as a bot where it like automatically posts it in the Discord, right? So maybe we can even set that up right now, as I would say.

• 33:07Speaker 01

  You know, create a Discord message on everything that was merged into the main branch. Yeah, into the like master branch.

• 33:23Speaker 01

  Of the code today. So is this, is this interesting to work through?

• 33:27Greg Isenberg?

  Oh, 100%.

• 33:29Speaker 01

  Okay, great. So we'll give this to, I don't know, the content strategist, I suppose. And this will be for the Paperclip app. App, here we go. So here's how I would actually do this. I would say, every day, we put work, put work into Paperclip.

• 33:53Speaker 01

  We're merging pull requests from the community and adding features. We want to have a Discord channel where we post the updates

• 34:09Speaker 01

  in a community, like in a community-formatted, like, form.

• 34:17Speaker 01

  We especially want to celebrate our community members who contribute. So make sure you call out everyone who added something that was merged. We do take the weekends off sometimes, so if there isn't anything, you, you can just do nothing. Okay?

• 34:51Speaker 01

  Read the GitHub changes for the last, last 24 hours for this. Okay, I know that took a little bit of time, but you know, the repo is here, so you can find the Paperclip repo here. It's all open source. We've got, yeah, 500 pull requests that we're working on. We've got 30,000 stars, so there's a lot of, like, motion happening in this repo that we want to be able to kind of celebrate every day. So.

• 35:25Speaker 01

  Fine, I think this is good, good enough. Like, oh, this is what I would say is,

• 35:32Speaker 01

  eventually we'll post it, we'll post direct to Discord. But for now, just post in your issue. Okay, so I'm going to create this routine, then from there I can do a trigger. I can say, you know, every day at 10 a.m. I add a trigger here. We can run it right now and we can see what was actually run. Now, this, a routine, right, is actually almost like a template of an issue. It's an issue that you're going to like rerun over and over and every day. But you still with Paperclip get

• 36:09Speaker 01

  that same kind of tracking where you're able to see for this particular day when we decided to run this, what happened? How many tokens did I spend? What did the agent actually do? What did you find out? So you, so you, so you get, instead of just like having some job that runs in the background that you have no tracing over, every single task that Paperclip does, you can go back and look and make it better.

• 36:36Greg Isenberg?

  Is Paperclip the most successful Paperclip, Paperclip project that you know of?

• 36:41Speaker 01

  Um, so one of the, so I guess that goes into the, the issue of like, so who's using Paperclip today and what are they doing? One of the things that we're finding is like, so Paperclip has been out not even three weeks. We're finding that a lot of folks that are having success with Paperclip are using it to help manage AI business, AI in their existing businesses, right? As much as the tagline is like run a zero human company, we're finding people who already have, you know, they already have a marketing firm, they're setting up agents. One person who's using it has a security review company and they used Paperclip to,

• 37:25Speaker 01

  to do their own security reviews on Paperclip itself and they're using it to help manage their clients for like automated security reviews. So, you know, we're, we just had a, there is like an, like a dentist just posted this week that he was using Paperclip to organize all the work that he has for a foundation that he's working on and managing his family. So, we're finding a lot of these kind of,

• 37:52Speaker 01

  Maybe untraditional long tail uses, like even like blue collar. Like I have a friend who he runs like a roofing company and they're, we're exploring using Paperclip to help their like sales agents find leads, right? So you have this idea where you want to find an area where there may be recently was hail and it's an area that like where they're probably going to have insurance, maybe a nicer area where they're willing to replace their roof or spend the money to do that, right? And so you can use Paperclip to say, well, go gather all the satellite imagery, go gather all, go gather all the hail data, come up with these, you know, create an app where you might be able to like give your sales agents a better chance at kind of closing the deal, right?

• 38:36Speaker 01

  So, so to answer your question most directly, I would say that like Paperclip is the most, has the most GitHub stars, but, um, well, Paperclip hasn't made any revenue yet. So I would say the companies that are making money that are maybe more successful arguably.

• 38:55Greg Isenberg?

  Yep.

• 38:57Greg Isenberg?

  Cool.

• 38:58Speaker 01

  So one of the things that we are shipping, um, probably by the time people are watching this is the ability to import and export companies. So, uh, for example, um, we have, I can show you this repo.

• 39:17Speaker 01

  Yeah, this will work. So for example, um, you've probably seen, uh, Gary Tan has his G stack, which is, um, a set of skills that he uses where you can do, um, office hours, like you're talking to Gary and his engineering and his style. You, we have, um, a tool which is not in the main branch, uh, but it will be, um, by the March 25th. Well, what you can actually do is install his company into your Paperclip, or like a form of his company, which will have all of the agents, all of the skills that you need.

• 39:58Speaker 01

  So there's other repos like this that you've seen. So the Superpowers repo or Agency Agents, right? This, I'm sure many of you guys have seen this repo. It has 60,000 stars, where there's some, over 100 agents. And you can import them all into Paperclip with all the same skills. And when it does this, it's not actually like copying the skills. It's actually referencing these remote repos, so you can import any upgrades. There's a game studio, yeah, the Don Cheetos Game Studio, where, you know, there's a creative director, producer, technical director. There's all sorts of skills around creating assets. So ahead of time, I imported this already.

• 40:42Speaker 01

  And I asked them to create a bullet hell game in Godot, inspired by Vampire Survivor, right? And so they start to plan the work. And I paused it because he was using too many tokens from me right now. But if you want to make games, yeah, you can basically have this massive game studio running on Paperclip. Now, the obvious question is like, yeah, but does it work? Right? Does this structure actually work? And I would say that like, it's just completely unproven at this point, right? Because there's never been a piece of software like Paperclip, where you could go to these repos that have, you know, 100,000 GitHub stars and 300 agents.

• 41:25Speaker 01

  And it's like, there's no evals for them. There's no like, there's no runtime. There's no runtime other than Paperclip to actually put these agents into an organization and test it out. And so that's really where the future is. If you, the future of like agentic programming is you download Paperclip and you create the organization that you've actually tested that actually works. It actually does create the TikTok marketing agency end to end. And then now you can share it if you want to, or just use it yourself, right?

• 41:56Greg Isenberg?

  Yeah, well, this is the sort of idea around, yes, I can go and create higher agents from from scratch, or I can go and hire a proven team, or acquihire a proven team and bring them into my Paperclip instance.

• 42:12Speaker 01

  That's right. That's right. So yeah, so that's Paperclip in a box. I mean, you know, I can answer questions that you have. I think that's, that's, that's where it is today. You can use any agent that you want. You and and really just it's open source. The community has a Discord server. I think that like, there's still a lot of rough edges, things that we want to work out around artifacts, having better onboarding. We want to have a, we'll add a CEO chat. So that way you're not creating issues for every single thing. You can just chat with any agent. We're working especially on my favorite feature that we're working on right now is called maximizer mode. And in maximizer mode, you basically

• 42:56Speaker 01

  don't really care that much about token spending. You're saying, I want to make sure that the CEO, the CEO makes sure someone is working all the time. Like, if I ask you to build the bullet hell game, you do whatever it takes to make sure that you have all the team that you need and that you're pressing on making it until that game is playable and you say it's completely done. So we don't have that today in Paperclip, but give us, you know, a couple of weeks and you'll be able to use the Paperclip maximizer.

• 43:26Greg Isenberg?

  When you say give us, who's us?

• 43:30Speaker 01

  Yeah, so us would be, there's myself, Dota. I also have two co-founders, Devin and Scott. And then also the community, who is just doing an incredible job at contributing just every day. We have so many poll requests, we can barely even handle them.

• 43:49Greg Isenberg?

  And is that Devin Foley, by the way?

• 43:51Speaker 01

  Yeah, absolutely. Yeah. So, yeah, he was at Slack early on, and at Figma early on. And then also Scott Tong, who was head of product design at Pinterest.

• 44:05Greg Isenberg?

  Cool. So it's not just an avatar. It's proven, proven founders. I know Devin from back in my days in San Francisco.

• 44:15Greg Isenberg?

  I think he actually, he invested in my last company. So I do know him and he's a great guy. So it's people who've built startups before. Scale them and it's cool to see what you all have built. You know, did you expect, I mean, I'll end it with this. Did you expect it to go this viral?

• 44:34Speaker 01

  I, you know, I wanted it to do well because I knew it was good and I knew it was something that I wanted, but I would say I didn't expect it to go as viral as it did. It really struck a chord with people. I think that like the,

• 44:47Speaker 01

  the big question in everyone's mind is like, will you need something like Paperclip in a year from now? Will the underlying models get good enough such that you don't need custom software? And I think that's an open question for everyone, really. Like it's not just Paperclip that would succumb to that. But I think that Paperclip is built to sort of survive this, this kind of, I don't know what you want to call it, the bitter lesson, which is that the models will always kind of outpace how humans do things. I think that like, if we're going to find that AI takes unfortunately, a lot of people's jobs,

• 45:29Speaker 01

  you still need a piece of software that helps manage the jobs that are left. And you still like, if, if, if I'm going to be able to not have 10 engineers under me, but I'm actually managing 10 AI agents or a hundred, you still need software that's going to help you manage the, the taste and, and the organization at scale. So I do think that Paperclip will have a bit of staying power because it's not fixed to one agent. It's actually operating at this higher level of just what do you want to do? And I don't know that like, what do you want to do will ever really change.

• 46:05Greg Isenberg?

  Cool. I'm, I'm watching. I think it's really interesting.

• 46:11Greg Isenberg?

  And you know, when it first came out, I hit you up right away. Actually, you hit me up. You're like, check this out. And I was like, whoa, this is crazy. And I'm happy you decided to come on the pod. Had to beg you, but you came on and I'm glad you did. This has been really, really interesting. I'll include links on where you can follow Dota, where you can sign up to Paperclip, play around. And thanks for coming on the Startup Ideas Podcast, my friend.

• 46:39Speaker 01

  Thanks for having me. Thank you.