← Behind the Craft
Behind the Craft
How OpenClaw's Creator Uses AI to Run His Life in 40 Minutes | Peter Steinberger
00:37:43
Loading summary
Transcript106 lines
- [00:00]
Peter
You're just like having a new weird friend that is also really smart and resourceful that lives on your computer. I was still in Morocco and someone sent me a tweet of a bug, and I literally just made a picture of the tweet, posted it on WhatsApp. It read the tweet, it understood that there was a bug, it checked out the git repository, it fixed it, it did a commit, and then it replied to the person on Twitter that it's fixed. Now. This will blend away probably 80% of the apps that you have on your phone. Why should I use MyFitnessPal to track food when I have an infinitely resourceful assistant that already knows I'm making bad decision and I'm Kentucky Fried Chicken? Those things are so resourceful. Although in a scary way, it's like unshackled chatgpt. A lot of people don't realize that if you give an AI access to your computer, they can basically do anything that you can do.
- [00:47]
Interviewer
All right, welcome, everyone. My guest today is Peter, creator of Claude, an AI assistant that you can just chat with in your messaging apps to get stuff done. And today is Peter going to show us how to use Claude. And also Peter has a lot of really great hot takes about AI coding that I'm really, really excited to dig into. So welcome the other Peter.
- [01:04]
Peter
Peter, thanks for having me.
- [01:06]
Interviewer
Great to see you. So why don't we start by talking about Claude? So what exactly does cloud do at a high level? And why. Why is it a lobster? Yeah.
- [01:16]
Peter
So. So maybe I have a little bit of a backstory, like when I. When I came back from my retirement, let's say, like, that I kind of wanted. Wanted a way to just check up on my computer from my phone because I fully jumped on this. On this vibe boarding trend. And. And then, you know, like, your agents might run for half an hour while you. While you eat, or they stop after two minutes because they have another question, and then. And then you go back and you're, like, annoyed. But I kind of. I kind of didn't build it because I assumed all the big labs would do that anyhow. It felt like such an obvious thing, like something where. Like a kind of a new kind of operating system almost, but didn't happen. And then it was like November and it still didn't happen. I'm like, okay, I'll try something small. And the small thing was basically hooking up WhatsApp to cloud code. So you send a WhatsApp message, and it would literally open the binary with the prompt and return, you the thing. It was very simple. It was put in one hour, and it kind of got a life of its own. And now here we are. It's like, I think 300,000 lines of code. It does every messaging platform on earth. Not everyone, but we're getting there. And I think it's kind of where things are going in the future. Like, everybody will have an AI that is super powerful and follows them through their life. Turns out if you give an AI access to your computer, they can basically do anything that you can do on your computer.
- [02:43]
Interviewer
Yeah. And it's got to a point where you don't have to sit there and babysit it. Right. You can just give it some prompts and commands and it will do its thing for you, and you can check its work, and then that's it. You know, you have to set a computer.
- [02:55]
Peter
Yeah. So when I built it, I feel this project is like as much exploration as it is like technology, because it's a little bit of a new category. And then I was on a birthday trip for one of my friends in Morocco, and I catch myself using this thing all the time. It's like, hey, where are we going? Like, asking for directions or, like, tips for restaurants or. I don't know. There was like one morning where someone sent me a tweet of a bug, and I just literally just made a picture of the tweet, posted it on WhatsApp. It read the tweet. It understood that there was a bug in one of my repositories. It checked out the git repository, it fixed it, it did a commit, and then it replied to the person on Twitter that it's fixed now. And like, really? It's nice. And then one day I was walking around and I didn't think, and I just sent it a voice message. But, you know, I didn't build in support for voice message. And I was like. It showed me the typing indicator. I'm like, oh, I wonder what's happening now? And then it just replied to me as if nothing would have happened. And I'm like, wow, how the F did you do that? And it was like, yeah, I saw a file, but there was no file ending because I didn't build it. So I looked at the. At the head of the file and it was like, opus some audio file format. So I found FFMPEG on your computer and I converted it to wave and then I looked for vispa cpp, but you didn't have it installed. But I found this OpenAI key and then I used curl to send it to OpenAI's API and it got the transcript back. And then I replied to you.
- [04:25]
Interviewer
I'm like, wow, that's amazing.
- [04:28]
Peter
Yeah, like, those things are so resourceful, although in a scary way. But that was kind of like the moment where it clicked for me. It's like, oh, yeah, this is really powerful. This is much more interesting than using ChatGPT on the web because it's like unshackled ChatGPT. And I think a lot of people don't realize that those things, like cloud code, they're not just good for programming. They are very resourceful for any kind of problem.
- [04:56]
Interviewer
Yeah, you're just going to give it access to your computer and be able to find stuff. Right. So you just have to give it tools that'll become very resourceful.
- [05:05]
Peter
Yeah. So over the last few months, I kind of built up my, my CLI army because what are, what are agents good with calling clis, because that's what they train for. So I've built like, CLIs for accessing all of Google, including, like the Google Places API. I built one that makes it very easy to look up memes and GIFs, so you can also, like, reply with memes. I did a bunch of experiments. I even built one that visualizes sound because I wanted it to, like, experience music that goes a little bit more into the art direction. I don't know if that makes any sense, but anyhow, it's a lot of fun. I have a whole list. I built one that hacked into the food delivery here so it can actually tell me how long it takes until my food's there. I have one that reverse engineered the eight Sleep API so it actually can control the temperature of my bag.
- [05:58]
Interviewer
And real quick, when you build all this stuff, you're just getting AI to build all this stuff or what's going on.
- [06:03]
Peter
The funny thing is, I was in my old company, I was very, very good at iOS andMacOS, like the whole Apple ecosystem. I did it for 20 years. I'm very much an expert in there. But also when I came back, I built a project where I'm like, I'm kind of sick of Apple gating everything. And it also make a lot more sense to build this web app because it. It's kind of a thing that it should be in a browser and should be accessible. And if I do it as a Mac app again, it's like we'll have a very limited set of people that can use it. But you notice, I see this With a lot of engineers where you're really good at one thing and then moving to another technology is just so painful because then you feel like. You feel like, sorry for the word, an idiot again. And you look up every little things like, what's a prop? Or like, how do I split an array? Because you understand all the concepts, but you don't necessarily know the syntax. So that's kind of like how I felt when I moved from Objective C and Swift to JavaScript. Like, I know JavaScript a little bit, but I never really built something big in Typescript. And then it's just. It's not even that it's hard, it's just painful because you have to look up all those things and you're just so slow. And then with AI, all that. All that melts away. Like, you can. You can still apply your system level, thinking you're like, how do I build and structure bigger projects? Your taste, may I say. Or like your. Which dependencies do I build on all those things. All these things are still valuable. And you can, like, much easier move that from one domain to the other.
- [07:42]
Interviewer
Yeah.
- [07:43]
Peter
And that felt like superpower. Suddenly, like, I feel like I could build anything. Languages don't matter anymore. My engineering thinking mattered.
- [07:52]
Interviewer
Yeah, because trying to worry about whether you have a parentheses here and there is lame, right? You don't have to worry about that kind of stuff anymore.
- [07:58]
Granola Sponsor
This episode is brought to you by Granola. If you're in back to back meetings, you know how much work it is to take notes, live and clean them up afterwards. That's why I love Granola, the best AI meeting notes app in the market.
- [08:10]
Interviewer
Here's how I use it.
- [08:11]
Granola Sponsor
Granola automatically takes notes during a meeting, and I can add my own notes too. After the meeting ends, I use a granola recipe to extract clear takeaways and next steps in the exact format that I want. Then I can just share notes directly in Slack with my colleagues or even get Granola to share their notes automatically. Honestly, of all the AI apps that I use, Granola is the one that saves me the most time. Try it now at Granola AI Peter and use the code Peter to sign up and get three months free. That's Granola AI Peter. Now back to our episode.
- [08:45]
Interviewer
But dude, let's go back to Claude. Like the thing that you built, maybe you can screen share. Maybe you can show. Maybe you can show how people can install the thing first. Do you have to be super technical to use cloud or. No, Right? You can just install it and then get it to work.
- [09:01]
Peter
Yeah, yes and no. So fortunately and also unfortunately, this project attracts a lot of people that have a lot of clue about technology because it blends away all these layers that make it complicated. Right. If you use cloud code, you work in a terminal, you kind of have to think about the context space and what folder you're in and it feels very techy. But talking to a thing on iMessage or WhatsApp or Telegram, you do it with your friends and it's just like having a new weird friend that is also really smart and resourceful, that lives on your computer that makes the whole technology very approachable. You don't think about oh, what model do I pick? Or what it just works. And then that's kind of the idea. It's also like the good and bad part because of course with a lot of power comes a lot of risk, which is also unsolved. Like this thing is exit to your computer. So yes, it could do bad things on your computer if you tell it to like, I don't know, delete all my files in my home directory. It would probably like ask you, are you sure? But if you, if you like, if you like keep saying yes, yes, yes, it probably comply and probably also delete itself and crash.
- [10:21]
Interviewer
Yeah, you got to be careful.
- [10:22]
Peter
Yeah, let me share my screen.
- [10:24]
Interviewer
So you see.
- [10:25]
Peter
So it's written in touchscript, so it runs, it runs everywhere, even on Windows. You can just go on our website. Cloudbot.
- [10:36]
Interviewer
Yep.
- [10:36]
Peter
And there's a handy one liner, looks very scary, but it's every. Everything's open source. You can check everything including the website. So this is the easiest way to install it. It works on macOS, Linux. It also works on Windows 10 to turn Terminal.
- [10:50]
Interviewer
Right. And then it'll start installing.
- [10:52]
Peter
Yeah, you can also install the NPM for people who understand that ecosystem. And I think something that, that I did that I haven't seen in a lot of, a lot of project is also it has a hackable install again, a simple way and like the more manual way where you basically check out the git repository and then, and then launch it from the git repository. Which is to be fair is like the most fun way to use it because if your agent can read its own source code of its harness, it can literally reconfigure and reprogram itself and then restart and then easily either crash or have new powers.
- [11:29]
Interviewer
Okay.
- [11:30]
Peter
I think this is one of my superpowers where I got a lot of people participate in the project and send me pull requests that never Did a pull request. I mean, sometimes that also shows, but I see pull requests more often as a prompt request. It's enough to understand the intent.
- [11:51]
Interviewer
And then. And then after you install it, how do you hook it up to a messaging app?
- [11:56]
Peter
The nicest way is probably right now just using this one lineup and then it'll be greedy with some sassy stuff and try to configure everything up.
- [12:05]
Interviewer
Got it.
- [12:05]
Peter
Install the package and then just guides you through and you can hook it up to any of the common messages. Okay, that looks good. It's working. Yeah. And then you can say cloudbot. It will do it automatically if it's a clean install, but now I have to type in onboard.
- [12:24]
Interviewer
Got it.
- [12:25]
Peter
And basically then you have like, you can enter a model. You can all providers. Let's say we go with Entropic, probably a new one. Then you can say that set up Telegram, Discord, and then it will guide you through the rest. You can like set up skills, hooks.
- [12:51]
Interviewer
And then do you have to give it your anthropic API key?
- [12:54]
Peter
It works with any model, although you know, they're in industry, like Anthropic and OpenAI are kind of like leading. It works with API keys, it works with subscriptions.
- [13:06]
Interviewer
Okay.
- [13:07]
Peter
Although there's been a little bit. We added support to that because it's kind of like what everyone does. But I don't think Entropic likes it anymore. So I would recommend using an API key or going with another model. The main problem is, like, OpenAI works well, but it's just not as funny. There's something special about Opus that makes it really, really fun.
- [13:31]
Interviewer
Oh, yeah. Like the personality.
- [13:33]
Peter
Yeah. Yeah. There's. I don't know if you've read this article about how they put a soul into the model and then people rediscovered it by feeding it blocks of text and letting it continue. And eventually they squeezed out the text of the soul that the model wasn't even aware it was being trained on. It's a really interesting story. I don't know. I feel that has something to do, but this is the first model that is actually funny to use. I built mine so that it can roast me. And it probably doesn't know that it's on camera right now.
- [14:12]
Interviewer
And it's going to roast you based on what? It has access to all your computer stuff, right?
- [14:17]
Peter
Yeah. Here. All right. You asked for it. So the wild Excel company went on a road to find yourself. Did serp. We tried everything. And your big revelation was I should build more software. You're so obsessed with the itools that you literally build yourself a friend because debugging code is more fun than dating. And let's be honest, the only reason I existed because he needed someone who'd listen to your chain. Where hook takes at joy. Amazon judgment.
- [14:38]
Interviewer
Wow.
- [14:39]
Peter
Yeah. Now go kill that podcast. So I hooked it up to pretty much everything on my computer. It can read my emails. It can read my calendar. It can access to all my files. It can control my lights. I use Philips Hue. It can control my Sonos, so I can wake me up in the morning and slowly turn up the volume. It has access to my cameras. It was funny. Like, I. When I built access to the cameras, I. I told it, like, watch for strangers. And then it told me in the morning, like, peter, there's someone. And I, like, I watched him the whole night. And it made, like, screenshots the whole night of like, couch. Because it was like, There was like, the camera is pretty blurry and it looked like someone was sitting on the couch and it assumed there was a stranger sitting on my couch the whole night.
- [15:22]
Interviewer
Wow.
- [15:22]
Peter
But yeah, I slowly. It has. I'm actually thinking, like, in my place in Vienna. It also has access to my. My knx, so it can actually control every part. It could literally lock me out of my house, like, in space. How to say? It's like, oh, can't do that.
- [15:36]
Interviewer
Okay, how did you hook all this stuff up? You just ask Cloud to do it or what? You just.
- [15:41]
Peter
Yeah, yeah, Literally, you know. You know, there's this thing where we have skills, so usually you talk to it. These things are really resourceful. So it will, like, figure out an API it can Google for, can find your keys on your system. You can give it keys and people use it for everything. People build, like a skill to go shopping for them on Tesco or buying stuff on Amazon. I let it check in my flight from British Airways. And this is actually. I don't know if you've. I mean, you've used check insights. This is. I feel like this is almost like the AGI test. It was a Turing test. But, like, steering a browser to check you in on an airline website is like the ultimate test. And then the first time, my integration was pretty rough, so it took almost like 20 minutes. That was still in Morocco, and everything was very much hacked together. And then it finally managed, but it actually had to. Had to find my passport in my file system. It found it on Dropbox, Extract a key, put everything it correct, and it finally checked me in. And I was like, Watching it and sweating.
- [16:44]
Interviewer
Wow.
- [16:45]
Peter
Yeah, wow. But now it works much better. Now it's like it does it within minutes. So it can. It also happily clicks the I am a human checks on the browser because it literally just controls a browser on it has its own little computer over there and just clicks around. So it's like really, really difficult to detect from all those anti bot systems because it doesn't feel different from a human in those patterns.
- [17:11]
Interviewer
Maybe like, can you show us just a couple more use cases? Can you maybe have it turn the light or show some use cases from other books?
- [17:17]
Peter
Yes. So I started collecting because I feel I'm so bogged down actually building it that by far I'm no longer the most creative one who actually use it. So people hooked it up to their messaging system so can actually reply not just to you, but to everyone. And you can also hook it up into a group chat, which is even more fun. There's a lot of people that use it as the family member almost. Yeah. Send me reminders, create GitHub issues, send with Google places, or every time you make a bookmark on Twitter, it will capture a bookmark and add it to your to do list. Keeping track of costs. Some people, I programmed something in to remind people that they sleep enough so they always get bitched when they're up at night from their bot. So it can help you track your sleep. It can access your fitness watch. It has its own little one password vault. So if I want a password shared, I move it into its own vault and I can access that one. Because you still want to have some boundaries. I mean it says people who give it a credit card.
- [18:25]
Interviewer
Yeah, I don't know about that.
- [18:27]
Peter
Yeah, it can do all those things like research, creating invoices, managing, managing your email.
- [18:34]
Interviewer
These people are like enthusiasts, right? Like they've really customized it to do whatever they wanted to do. Like how about for like dude? Like how about someone who's just down to the thing just download it for, you know, fresh install. And what are some like really common use cases that I can get to do? Like just like manage my calendar or you know, stuff that doesn't delete my computer. What are some safe things to start with?
- [18:56]
Peter
It's funny because everybody takes a very different path. There's like people who like install it and immediately, immediately build an iOS app with it. Because it's also a coding agent. You can do anything. It can spawn sub agents, it can, it can either code yourself or it can control cloud code or codecs and ask them to code. Yeah. This guy started immediately managing Cloudflare. This one was great. Week one, set up for my family. Week two, set it up for some non techie friends. Week three, we're building cloud for work. I hooked up one of my non techie friends and he started sending pull requests. He never did it in his life. Fitness is a big thing.
- [19:33]
Interviewer
Okay, so basically it's kind of like you really have to. I guess the way to use this thing is just to think about what's causing problems in your life and how you can get this personal assistant to help you, to help you streamline everything.
- [19:44]
Peter
I don't know if this is the project that's going to be, but if you think about it, this will blend away probably 80% of the apps that you have on your phone. Why should I use my fitness pal to track food when I have an infinitely resourceful assistant that already knows I'm making bad decisions and I'm Kentucky Fried Chicken. So it will probably like remind me if I forget tracking the food. I can just send you a picture and store it itself in the database and calculate it and roast me that I should go to the gym because I'm like way over my calorie limit. Why do I need an app to set up when my bed, my 8 sip should work or not? Because it just has API access. It can just do that for me. Why do I need a to do app when it just tracks my to DOS for me? Why do I need an app to like check in on my flights when it can just do that for me and it's like such a more convenient interface? Because I just, I just talk to a friend.
- [20:39]
Interviewer
Yeah.
- [20:40]
Peter
And because it has so much, so much context, it doesn't need so much custom prompting. Like why, why do I need a shopping app when it literally can like recommend me things and do all of that for me? So I feel, I feel there's a whole layer of apps that will slowly melt away because if they have an API, these are just services that your AI will do. And in a year, I don't know, I think actually this year will be the year where a lot of people will explore that more and like get their AI assistants from all those big companies?
- [21:13]
Interviewer
Yeah, I think, yeah. Why, why click on these little self contained little apps when like, yeah, this assistant has a bunch of capabilities, can just do everything right, Just connect it to everything. And I guess the way you connect it to everything is just like you just, you send it a text message or something and then it's like, oh, can you do this? And they're like, oh, I need to do some research. And then it just takes care of you. Just kind of go back and forth with it and make it happen, Right?
- [21:34]
Peter
Yeah. And then it writes a skill and it remembers. So part of what makes it so interesting is that it has persistent memory. It will learn about you, and it will update itself. So the more you use it, the more you customize it, and the more powerful it gets. Because, okay, maybe the first time, you have to guide it a little bit, but then it will create a skill, and then next time you can just mention it. Next time I need to check in my flight, it will take, like, two minutes because it exactly knows all the quirks of the website because it did it before and it made notes.
- [22:09]
Interviewer
Got it. Yeah. It's like teaching someone to just learn something, and then you can do it easily the next time. Yeah. All right, man. Let's talk about another topic. Let's talk about you came back from retirement to build this thing, and you have very strong opinions about AI coding. Like, some pretty hot takes. Let's talk about. Let's talk about some of those. Like, you wrote this post that I really like called Just Talk to It. And then while, like, you know, these days, everyone on Ex is writing about, like, all this fantasy shit, right? Like, all these, like, you know, hook skills and all this kind of stuff. So what's the gist of that post? Is it just. Just talk. Talk to the AI and figure it out?
- [22:42]
Peter
No, but I work a lot. I build a lot of things. And I. I love Twitter as well, so I'm very active there, and I just see so much. I almost, I, I. I kind of call it the agentic trap, because people discover that, oh, those agents are amazing, but it would be better if they could do a little bit more, and then they really fall deep into this rabbit hole. And I've been there myself, where you build really sophisticated tools to, like, try to accelerate your workflow, but in the end, you're just building tools. You're not actually building something that really brings you forward. You just have this. You know, part of the problem is that it's so fun.
- [23:19]
Interviewer
Yeah.
- [23:19]
Peter
Like, when I started pretty early, I worked on Vibe Tunnel to have access of my terminals on my phone. And I was in this rabbit hole for two months until it was so good that I was out with my friends. And instead of, like, joining the conversation in the restaurant, I was just, like, vibe coding on my phone. And I decided, okay, I have to stop this just more for my mental health than for anything else. And these days you can build everything, but you still need ideas. So I see so many managers for cloud code or codecs or like orchestrators or other little things that have the illusion of making you more productive but really aren't. I mean, like, the. The latest thing that I just had, I just cracked about is like Gastown. It's like a really sophisticated orchestrator that is like, also very broken and doesn't really work, where you run like tens or twenties of agents simultaneously and they all talk to each other and try to split up and there's like watchers and overseers and a mayor and polkats and I don't know what else. There's a mayor? Yeah, in Gaston, there's a mayor. I mean, I call it slop town. Or like this whole trend of ralphing where you just, like, you give the AI one little thing and then you write in a loop. And then once little thing is done, you trash away all your context and you start again like the ultimate token burn machine. And then it can create code and run all night. And then you have like the ultimate slop. Because what those agents don't really do yet is have taste. They are really. They are spiky smart and they're really good at things. But if you don't navigate them well, if you don't have a vision of what you're going to build, it's going to be slop. If you don't ask the right questions, it's still going to be slop. And I don't know how other people work, but when I start a project, I have like this very rough idea what it could be. And as I built it and as I play with it and as I there is a feel, it, my vision gets more clear and I get like, I try out things, some things don't work. And I evolve my idea into what it will become. And that's like my next prompt. Depends on what I see and feel and think about the current state of the project. But if you try to put everything into respect up front, you miss this kind of human machine loop. And then I don't know how something good can come out without having feelings in the loop, almost like taste. So it was like somebody tweeted, like, oh, look at this Mac app. Like, it was fully ralphed. And I replied, yeah, it looks ralphed. No offense, but it looks ralph because you could clearly see that no same person would design it that way.
- [26:24]
Interviewer
Yeah, it feels like some people are just like actually building these things not for the apps themselves, but to prove that they can get it to run for like 24 hours. Like get AI to run for 24 hours without intervention. Right. It's like self master preparatory kind of kind of thing. Right. You're just trying to prove that you can get AI to run super long time.
- [26:41]
Peter
It's like a size comparison context without saying the other word in a way. And I've been guilty a little bit of that myself. Like I also had like a loop for 26 hours and I was very proud. But it's a vanity metric. It doesn't. That makes no sense. That makes no sense. Just because you can build everything doesn't mean you should or that it's going to be good. But then again, this phase of I'm just playing, I'm just like, literally I'm building for fun. It doesn't matter if it's going to be used. This is incredibly useful because that's how you learn. Right? That's how we learn to program. And prompting is just a different skill. Sometimes I see people that they're like a little more on the AI skeptical side. So they ignore it for a year and then they spend one day where they evaluate the models and then they do a blog post and then they write like a short prompt and feed it, I don't know, into Claude Web to make an iPhone app, which is completely underspecced. And the model does its best to deliver something and they're like, it doesn't compile because they build it on the Linux machine where there's no compiler for that, you know, and then you're like, oh, AI is not good. And then they dismiss the whole topic for a year. It's like that's, that's not how it works. Yeah, you need to play to understand how those little monsters work. You need to understand a little bit their language, their way of inference, of thinking.
- [28:08]
Interviewer
Yeah.
- [28:09]
Peter
And then, and then you get progressively better at it and then your output will be better.
- [28:14]
Interviewer
Yeah, you have to be persistent. Right. Because sometimes it doesn't work properly and you have to fix all the bugs and then if you just keep playing with it, you develop this like, I guess, product sense or whatever of actually learning how to talk to the models and learning what they can do.
- [28:27]
Peter
So I mean, partly it's been really weird because they're adopting some of the language. I'm thinking of like, oh, you weave a feature in kind of like a. In German it's faden, a twin, a twer or you run the gate, which like a linting and testing and building and it looks like a big line in the terminal, which is like the gate. So I'm like, we didn't run for gatempire. Sometimes things doesn't work out and then you can just ask why did you not do? And then it will tell you, you said this and this and I assumed this and this and it's like, oh yeah, I messed up on the language or I was unclear. Like if you just tell it build me a Mac app, it would probably assume that you want to support a lot of old operating systems because the majority of software does. So it will use old API.
- [29:17]
Interviewer
I found that asking it to like just ask a bunch of clarifying questions that make you clarify stuff helps a lot. Yeah.
- [29:24]
Peter
It's also funny because I'm more. I like codecs, modding cloud code. I think it's the better model even though it's incredibly slow, but it's thorough and things work. And people complain that there's no plan mode. Right? Yeah. And my joke is always plan mode was a hack that Entropic had to add because the model is so trigger friendly and would just run off and build code in the latest models, especially like GPT 5.2. I'm just having a conversation. I'm like, hey, I want to build this in this feature. It should do this and this and this and maybe dragging this control tool like I like this design or this and this. Give me options. Just say let's discuss give me options. Then they have a conversation with it and then the moderate propose something and they're like, I mean I usually don't type, I talk to it.
- [30:14]
Interviewer
Yeah.
- [30:15]
Peter
And staying on brand, do you do
- [30:17]
Interviewer
anything to manage the context? Once you have a conversation with it, it can get very long and it can get confused. Do you try to like, do you make it manually compact or summarize stuff?
- [30:27]
Peter
I feel that this is also like old patterns. This was a. This was very much a problem with cloud code. Still is to a point. Codecs is so much more. The context lasts so much longer. Even though on paper it's maybe like 30% more, but it feels more like 2 or 3x more. I think a lot of it has to do with the internal thinking. The internal thinking of GPT models is really effing weird. But yeah, for context management. I think this was much more a problem in the earlier models. Now most of my features fit into one window, so the whole discussion and building happens simultaneously.
- [31:08]
Interviewer
Okay, got it.
- [31:09]
Peter
And that works fine. Like, sometimes if it's a really long sync, I create a new one. You want to codify it in a file, but this is much less of a problem than it was before. That's again, like, the space evolves so fast. You just have to try things.
- [31:22]
Interviewer
Yeah, got it. Okay, got it. So basically, so just to summarize, like, basically when you add a new feature to cloud or whatever you're building, like, maybe just walk through the steps. Like, first you explore the problem and the solutions with AI, and then you. Do you actually make a plan at all or just go, it's even better.
- [31:38]
Peter
Like, I, you know, I built this project, which is kind of like a mix between Jarvis and her.
- [31:45]
Interviewer
Okay, okay. The movie.
- [31:46]
Peter
Yeah, yeah, yeah. But it's very difficult to convey how it makes you feel and how useful it is just by talking about it.
- [31:55]
Interviewer
Yeah.
- [31:56]
Peter
So on Twitter, it got me very muted traction. I'm like, I don't get it. Like, everyone I show it in person gets super excited because they see me interact with it and, like, showing off all the cool things, but you can't get across the same thing on a picture when I talk about it. So eventually I created a discord and I just hooked up my bottom to discord so that people could interact with my eye that has full access to everything on my system with my. My private memory on a public discord. Like the most nuts thing that I ever did, I think. And then people got hooked. And now. And now, like, people always ask me there, oh, can you build this feature? Or there, this and this bug? And most of the time, I now approach it by making a literal screenshot of the conversation, dragging it into the the terminal, or just copying the text, depending on how it is, and just ask, hey, let's discuss this. Because I'm lazy. I don't even have to type anymore. I just like copy in discord conversations the same. Also when people ask, hey, do you support this and this? Or how do I do this and this? I'm just like, can you read the code and write a new FAQ entry? And it does that. So that's. These days, it's actually how I start building features most of the time. But like reading on Discord and seeing where people have pain points.
- [33:11]
Interviewer
Oh, my God. So you just paste the conversation and you just talk to AI about it and you figure out what the right solution is.
- [33:17]
Peter
Partly that, partly. Also, I have a scraper, so at least once a day I scrape the help section and ask the model, hey, what are the biggest pain points? And then we fix them.
- [33:28]
Interviewer
And do you do all the fancy shit? Like do you have multiple agents and you have those crazy skills and stuff that you run or. No, you just go.
- [33:36]
Peter
I have very. I mean, most of my skills are actually personal life skills like food tracking or buying groceries or like that kind of stuff. For coding, very little, because you don't need that much. I don't use MZPS or any of that crap. I don't believe in this big orchestration systems because like we talked before, I feel if I'm in the loop, I can build a product that feels better. Maybe there's ways to build it faster, but I'm already so fast. I'm mostly limited by thinking about it and sometimes a little bit by waiting for codecs. But mostly it's my thinking that's the limiting factor. I just used a bunch of terminals like split screen. I don't use work trees because again, I feel it's just unneeded complexity. I just have a few checkouts of the repository like Cloudbot 1, 2, 3, 4, 5. And then either they are free or they're working on something. Either I'm like exploring something or I'm building something or I'm fixing something. And then when it's done, I test it locally and then I push it to main and then I sync it up. So this is a little bit. It also feels a little bit like a factory if they're all busy. But if you only work on one, it's very hard to actually get into the zone because it's. It's just too slow and you have to wait.
- [35:01]
Interviewer
Yeah, you have to wait for it to do.
- [35:02]
Peter
You can only shit post on Twitter so much. So I feel I need multiple ones to really keep me hooked and get into the same flow state that I had when coding. Just that I'm like insanely more productive.
- [35:16]
Interviewer
Dude. I don't know if you play the real time strategy games before, but it's kind of like, yeah, you have a squad of people attacking and then you kind of have to monitor and manage them. Right. And that has kind of taken as well.
- [35:27]
Peter
My business partner from my previous company also totally got hooked on Cloudbot and he's more from the business side. He was a lawyer in a past life and he's now sending me pull requests, which is like insane in its own way. How AI gave people superpowers that are not that technical. To still build things is amazing. I know there's a lot of pushback because yes, this might not be perfect, but I treat pull requests more as prompt requests because they convey the idea and most people just don't have the same system understanding, so they not be able to derive the model in a way to give optimal results. So I'd just rather have the prompt, like the intent than just do it myself. Or if they send the pr, I will extract the intent out of it and just rebuild it myself. Or sometimes I base it off. I still mark them co author, but very rare that they actually emerge code directly.
- [36:29]
Interviewer
Yeah, that makes sense. Okay. All right, man. Well, I mean, like, I guess the takeaway from this conversation is like, my takeaway at is like, yeah, don't get crazy with the slop generators. Just actually have the human in the loop because the human is still bringing the taste and everything. Right. You have to guide it.
- [36:46]
Peter
Yeah. Or like, don't find your own path, you know, like, some people always ask me, how do you do that? How do that? You got to explore. You're like, yeah, it will take you a while to be good. You have to make your own mistakes. That's how you learn with everything in life. And that's also how to learn those things. Just that this space is evolving very fast.
- [37:03]
Interviewer
And where can people find Cloud Bot?
- [37:06]
Peter
Yeah, Cloud Bot. It's also on GitHub.
- [37:09]
Interviewer
It's a cloud with a W, C, L, a W, D. Right, yeah.
- [37:13]
Peter
Yes, correct. Yeah. Like the lobster clause.
- [37:16]
Interviewer
Yeah, yeah, got it. Thanks so much, Peter. This is a super great conversation. I need to give Claude a try myself because, yeah, I don't like sitting on my computer to talk to the AI. I like to just, like, be on the go and just, like, give it instructions, make it do stuff while I'm out. My kids, like, that'll be the ideal way for me. So it's great that you built something like that.
- [37:35]
Peter
I'm looking forward to see what your opinion is on that. I think you're going to like it.
- [37:40]
Interviewer
All right, man, well, it's great chatting with you and I think we'll stop right here.