How to Be "Agent Native" in 2026 w/ Every CEO Dan Shipper

33:53and you know i had a couple of experts look at this over the last couple weeks because one of the one of the fun things about doing this stuff so publicly is like when i when you tweet and you're like hey like my thing is down um people who have a lot of experience with ygs come out of the woodwork and be like hey like i could take a look and what's really cool is like because i like sent them the the repo and then ducked you know because i was just like this is like they're gonna judge me so hard yeah yeah yeah like i was like i'm so sorry for this um what's really interesting is they're all like yeah this is actually very reasonable um it it lacks some amount of coherence so you can you can see that the agent was like solving local problems in a particular way but then not zooming out and being like well i solve it like this over here and like this over here and they those should match so i can understand the whole thing like it wasn't thinking like that which is what a good engineer would do i actually don't think that that's a permanent thing i imagine that would get better over time i actually want to ask you again on the

34:57Corey Noles:problem yeah do you think that that's a context uh window limit thing like where it just can't possibly think about the whole project at the same time or it might be it's also like a prompting thing i'm sure if i like prompted it a little bit better uh it would be it would be a bit better and also just honestly um it makes it better to do these kinds of projects if it's a production app if you hold in your own head the basic way that the architecture works. It's just better. You have to remember this thing is a super intelligent thing that pops out of a box every time you prompt it and it doesn't know anything for the last year and it's never seen your project before and it has to get up to speed every time.

35:46It's just hard. you know um so an extra 30 seconds to really explain what you want clearly can probably save you a lot of grief on the back yeah exactly explain what you want and also part of that is knowing what you want and especially if you've vibe coded it like you may not fully know um and so uh basically like i had someone come in and help me uh just okay just be like okay if we went back to first principles how would we architect this and i guarantee like i already knew most of what he said it just like wasn't fully there because i was like trying to transition from i didn't even know that codex didn't know the best practices to okay i'm having codex do the best practices but it's still like slightly doesn't want to do the full like rewrite and delete a lot of code and whatever and the guy that i brought in who's super talented was just like yeah this is this is exactly the thing that we need to do and like i'll just rip out a lot of the code and he used codex to do it but wow um it's not it doesn't necessarily come naturally to the ai models to do this yet um and now now it's like fairly stable i'm like happy with it you know some of the like some of the code is different but it's not it's not like a totally different app than it used to be and it happened very quick like he was able to essentially stabilize it in a couple of days of work um so it's pretty crazy what you can do and it certainly at the scale that we're at and and the level of sleepless nights that i was having um i probably would be more careful next time but i think generally this is fine uh just curious did you use codex's plan mode first i was using codex's plan mode yeah okay i was like i've had really good luck with planet a lot of times i'll even go and use a different ai and be like okay i'm going to workshop this idea uh you should know that i don't know what i'm doing entirely i have i have ideas and i can understand it if you tell me but don't assume i know anything yeah and usually i can get like a good here's the project i want to do prompt that has saved me some grief but it's it's always a coin toss you know you just never know and uh i i think you also kind of hit on that element of this thing we've kind of seen since since this all began is that in the hands of an expert who knows the right questions to ask uh it can do a lot more yeah it really can it really it contains all of the knowledge of all of humanity and you're only like kind of getting a little slice of what it knows you know and based on what you know to ask it yeah exactly um so it's a real skill to use these things and i don't think that's going away No, I don't either.

38:45Corey Noles:Do you think prompt engineering as a skill is worth investing time and energy into still? Or is it more like if you talk to it long enough and you give it enough information, you'll get what you want? I mean, I just think a prompt engineer – I think everyone is sort of a prompt engineer. But it's not like there are those little tricks that are like I'll pay you$2 ,000 or whatever and that you don't have to do anymore to get better results. There will probably always be like certain things you can do to like make it better. But the big thing is knowing how to manage the model, knowing how to ask for what you want and know if you're getting it back.

39:28And that's like kind of prompt engineering, but it's very specific to your workflow and what you want and the kind of thing that you're doing. And so I think it's prompt engineering is like maybe like writing. You write in all these different circumstances, and you can get better at writing, but we're really talking about specific things that you're trying to get done. So I don't expect people will study prompt engineering, but I expect that they will know some basics about certain little tricks, but mostly how to do it well for their specific use cases. It's almost as much about problem framing as it is about prompt engineering in itself and just understanding.

40:14I think prompt engineering has a role as like, here's step one for normal people. I think when average folks are like, I need to learn how to use AI, they can get a lot of unlock just from a quick prompt engineering course on Coursera or something. You really can, if you're new, really unlock a lot of capabilities you didn't have prior. But it's very much the starting point and not the end zone anymore, I would say.

40:40Corey Noles:I agree. We'll have one more question about proof. How well do you feel that you actually understand the code now? Now? Yeah. Do you feel like you really don't understand it? I still don't, but that's because I hired someone who understands it. Okay. Cool. Somebody does, though. That's all that matters, right? Somebody knows. That's how I would modify this. We had a little bit of a retro at every about this whole thing. When we launch something, sometimes we label it as an experiment, and it's okay for it to be a little rough around the edges. But if we're really launching something, we want it to be good.

41:14And also, on the other hand, when we launch something, I don't want to have to be up all night seven days in a row trying to fix it for my own health. Yeah, for real. so what i realized is we need a buddy system where if i'm doing this i need someone else who knows the code base and like knows a little bit about it to so that when we launch it if there are problems like it's not just me in like my foxhole you know trying to like understand the code base while it's going down and everyone's looking at me um because that's you can like switch off you can switch every other night exactly and that um and that led me to this articulation of how early product engineering teams should work which is the pirate and the architect model where okay you want a pirate and that's i'm a pirate which is like you're just going you're just going as fast as you can you're just trying to find something that like works and people like um and then the architect is a little bit like i want to really understand how the whole system works and make the whole system work together well as a well-oiled machine and especially in early product work you actually don't need a full-time architect i think you just need you just need a pirate just like going hard and an architect coming in for a couple hours a week to be like here's how all the things work and here's here's how we can tuck in the edges a little bit so that the core of it is stable.

42:42But you don't really want someone spending all their time making it perfect if you don't know if it's any good and you want to be able to explore as fast as possible. And sometimes some people on our team are kind of both. They can kind of flip in and out of pirate and architect mode. And I'm just like, not that. I just am not careful. Some things I'm very careful about, there's some not.

43:02Corey Noles:You also have a lot going on. I mean, you're CEO, you're managing the whole company. Yeah. So I think that's a good model. And I would expect to see more of that. And we definitely see that across. We've run five or six products internally. And we definitely have one person who's fully responsible for it all the time. And then they often have one or two people who are spending some part of their day on it, helping them with some of the big, difficult, more architect-y tasks. Yeah. That makes sense. so I think this is a good transition into agent native architecture um the second you published this I fell in love with it I think it's an awesome way of of of thinking about building applications to ride the or surf the models as you said uh and it was cool to see your interview with uh Mike Krieger uh from Anthropic yesterday and he said he actually uses it as a skill yeah um I don't know if you I I would have geeked out if I were you knowing that it was awesome I I was like, ah!

44:00Corey Noles:Yeah. Yeah. But yeah, could you just tell us kind of what that is? And if people are interested in building with agents, how they could apply it themselves? Yeah. There's a new way of building software. I've been calling it building software that's agent native. And it implies a new architecture for how your software works. And the way that you can think about it is normally in software, any piece of software is like a recipe. It has a set of steps for how it works, and those steps are known beforehand by the programmers. In this new version of software, instead of the whole thing being written out beforehand, it's essentially cloud code in a trench coat.

44:41It's like you take cloud code and you put some nice buttons on top of it so that you're interacting with a UI that feels familiar. But when you press the button, it sends a prompt to the agent, and the agent gets the work done. as opposed to running the recipe to get the work done. And there's a lot of really interesting effects of that. In particular, the interesting thing about agent-native software is anything a user can do, the agent can do. So if you can push a button that prompts the agent, the agent's going to have to be able to do anything in the app, and that's super powerful. uh another thing is that it creates this flexible way of working where the programmers don't necessarily know all the things it's going to going to do off like when they when they release it it's going to do things they don't expect so i think i think cloud code is like the canonical agent native application it's an agent it sits on your computer it has access to everything on your computer so anything on your computer you can do and it is it works in this way in this flexible way where it can run any bash command on your computer.

46:00So people were using it for code, which is the thing that it was intended for, but then they started using it for everything. It's like, organize my files, or plan my schedule, or whatever, and they're like, oh my god, this is so cool, and then they made co-work. And so it creates this much more flexible model of software that, it doesn't mean that traditional software doesn't work anymore, like I think the whole sass is dead thing is like such bullshit but there is this new class of software that um i think is really powerful and you know proof is is an example of this kind of thing there there's there's different types of agent native it can be agent native in the sense that it has an agent at its core internal to it or it can be agent native in the sense that all agents can use it natively so like figma at this point is agent native because it has a cli so there's a there's a whole new world of how software might work and also how software works with agents that it starts to open up.

46:59Corey Noles:Does this apply? Would you say that this also applies for people who are like building an agent to help them in their workflows? Like not necessarily building software, but like trying to, you know, whatever tool they use where they're having an agent deploy in production, or is that a separate problem? Give me an example. so let's say you're someone who's not technical and you want to build an agent to help you whether it's open with open claw or i guess actually open claw would be a good example like you have an open claw set up you want to have your open claw do something for you say like help you manage your schedule um would this apply in that circumstance too or is it more specific to you're building a software that people are going to use it definitely does apply but i think that open claw they've already built it to be agent native and so you're kind of getting the benefit of riding along with that architecture so an example of what makes it agent native is open claw is built on pi which is a like agent harness and pi is very very basic it doesn't really have much except an agent loop and the ability to modify itself and open claw puts a couple things on top of that so it has a cron job so it has a heartbeat so it it like wakes up every 15 minutes or so it connects natively to a couple of messaging apps but that's really it and the core of open claw is still this thing that can modify itself and that means it's super flexible right like um peter who built open claw he didn't i use it for bug tracking and and triage like he never built a bug tracking and triage feature into it and the the guy who made pi never built a bug tracking it's not made for bug tracking and triage but it's just flexible enough and its tools are granular enough that it can be used for anything and that's the interesting part of it that that level of simplicity is is what led to what is arguably one of the most transformative softwares we've seen is kind of awesome.

49:09Totally. It really is. And it opens up a way of thinking that is quite different from how programmers normally think. And it's actually hard to get AI to think this way

49:20Corey Noles:because it's trained to think like a programmer. And what programmers want is they want to be able to predict what's going to happen. Like they want to make a machine where they know how it all works. And I think that's one of the reasons why it took a long time to get something like cloud code is because we were pretty afraid to unhobble the model that's what anthropic talks about internally is unhobbling the model we kind of like really locked it down and we were kind of like oh it's going to be in this very specific type of workflow that it's going to work for and the real answer is actually give it a basic set of general tools and let it run in a loop and people will figure out how to use it for whatever their specific use cases are and i think that's it's just a it's a new way of thinking about software that's both scary and extremely useful you know and the way most of this is built in in you know big big research labs with with largely by software engineers i think there are a lot of capabilities that get skipped over that maybe they don't even know are there in some ways by just simply not taking that approach i always tell people ask it something you don't think it can do Always ask the thing.

50:33Some usually it will surprise you at how close it will get. Yeah, I totally agree. And I think that's why some of the stuff we're doing, the big model companies appreciate it and pay attention to it. Because they do have apps that they work on internally. But I also think of them a little bit like oven makers. So they're making an oven. You can use an oven for a lot of things. And we make soufflés. And so they make a new oven. and they come to us and they're like, tell me about the souffle you can make, you know, because that helps them figure out, like, how do I make the oven better? But they're not going to make it.

51:05How was the temperature?

51:07Corey Noles:Did it rise enough? Exactly. They're not going to make an oven just for souffles, but it takes them seeing it being used in a particular context where someone's, like, pushing it as far as it can go for them to even realize, oh, there's an opening here. There's a vector along which I want to improve it. And I think people don't quite realize that and quite realize the difficulty and also promise in making tools that are so general that you can't fully predict how they'll be used and how to improve them. So the dominant startup metaphor for the last 10 or 15 years has been jobs to be done. You have to narrow in on a specific job that your customer needs your model for or your product for and then make it better.

51:53And if you ask, okay, what problem does AI solve? It's like, well, it solves every problem. Theoretically, it could solve every problem to better and worse degrees. And that's really hard to figure out how to improve a product that's really meant to do everything. Yeah.

52:12Corey Noles:Yeah. Well, first of all, I can't go any longer without addressing this before we move too far past the pirate thing. JD Burrow in the chat said, I'm late to the party, but is Dan's last name truly Shipper or is he just emoting pirate vibes? I didn't even really think about it because usually people talk about shipper as shipping code. Yeah. But I do love the shipper as pirate thing. I didn't even thought of that. It is truly, truly shipper. But I'm just trying to make it make sense in every way possible. That's right. I love that. I saw that in the chat. I'd been eyeballing it too. Had to sneak it in there.

52:51Corey Noles:I want to get to compound engineering as well. Dan, how long are you here for? Are you here until 11 or can you go longer? I can go a little longer. Okay, cool. I just want to make sure we get to everything. So someone else in the chat asked, Gravemaster said, what's the first go-to get started with agents? Is it usually VMware clawed instances or something else? Well, you really teed me up there. You should try a plus one. If you want to get started with agents, we have a new product on every.to slash plus dash one, and it is your very own hosted open claw instance. you can get it with one click it has all of our uh every apps on it to help it do email to help it write well um you should you should really check it out it lives in slack it has all the right presets other than that yeah i think uh the vmware one is pretty good there's i think there's one for railway it depends on your particular setup and your particular thing that you want to use it for the i like having a mac mini i think that's kind of cool but i think they're sold out now.

53:57Definitely in Silicon Valley.

53:58Corey Noles:Where are you guys based, actually? We're in New York. I'm in Brooklyn right now. We never talked to anyone on that side of the country. I bet. Yeah, I bet. Cool, cool, cool. I wanted to ask one other question. Actually, let's just get into compound engineering because I think that would be kind of cool to talk about. We mentioned it very briefly before. You guys have a plug-in for this so people can actually use it whether they're using it with their Open Clause are their coding agents um can you tell us a little bit about this and i'll add a little bit of context um kieran was like the first um who's an engineer at every was the first person i saw besides like faceless people on twitter who was like maximally going hard on multiple coding agents and sub agents and like he was he in my opinion led the way on a lot of this stuff and uh ever since and he came up with this framework but but ever since i saw him i was like oh this is like the new way to do things um he he's a he's a true trailblazer um and one of the few i think senior engineer types who are like willing to give up coding like manual coding even before it was even before it was obvious it was going to work yeah and i think basically what like i remember very clearly there was some model some model came out i think it was claude opus 3.7 and we were testing it before it came out and when when we test models me and kieran often are like on a video call together and just like chatting back and forth and he was like i don't i think this just work like i don't think i have to look at the code anymore and so we were like trying that and we were like holy shit this is crazy and this is like maybe it's probably almost a year ago now um yeah and that then filtered into the rest of our company and we started being like i don't think we need to look at the code anymore and that became a thing that we were doing but everyone else was like that's crazy um and now it seems you know pretty much like that's the case pretty normal now yeah it's pretty normal it's pretty normal um And out of a lot of his experience with that came compound engineering, which is the idea that in normal engineering, each feature you build makes it harder to build the next feature.

56:26Why is that? Because the code base grows in complexity. All the complexity is interdependent usually. You have all these tests. You have a bunch of stuff that all depends on each other. Even in a modular code base, it still is like that. And in combat engineering, what you're trying to do is make each feature easier to build than the last. And that is by, as you do things, you compound the learnings from each feature into the next one. So like each bug that you find or each issue that you make, or if you're proof, like each first principle of building YJS applications that you miss, you like compound that into a research into a knowledge into your knowledge base in your repo so that every engineer has access to this um so and link and share it in the chat so yeah so the plus one link is uh let's i'll just put it right in here oh if you've got it yeah go for it yeah that should work thank you yep um so uh i'm just making sure that was right okay so yeah so that's That's basically how it works.

57:31It's a plugin. It's a philosophy, but it's also implemented in a plugin where there's four steps to it. The first step is planning, where it takes into account all the best practices, all the stuff you've learned in building your product, all that kind of stuff. It makes a really, really detailed plan. Then you kick off your agents to work on it, and often you do that in parallel. so you have a bunch of agents all working in parallel then you review or assess what happens so you you have tests and um you maybe test it manually you maybe have a fleet of agents do testing and then you compound that learning so you take everything that you learned and push it back into the first step of the process and that's what make it makes it compounding and i think this is something we were we were on early this is something that kieran in particular like really noticed and was like, I'm going to, I'm doing this.

58:19And I was like, that's sick. Um, and, and I think has become, even if it's not called compound engineering has become like a standard way that a lot of the model companies think about building engineering, like harnesses and doing programming. So it's, yeah, it's pretty cool.

58:36Corey Noles:That's really, yeah, that's awesome. Reminds me of how you and I test models. Grant, me and you. Oh, how, how so? What do you mean? Sometimes it's hopping on a live stream and throwing in stuff to see how it goes. Well, sometimes it's actually live stream. I was just meeting a Google call or something. Yeah, I got you. That's really cool. Well, I remember there's a way to do compound engineering with just being a regular anthropic user, and it's with skills. I don't know if you do this, Dan. But basically, whenever I learn that the model can do something that I'm often trying to do myself, I will turn it into a skill.

59:16Corey Noles:So then I don't have to ever prompt it to do that again. I can just say, you know, do this or in the off chance that it doesn't know to use the skill, I'll say, use X, Y, Z skill to do it. Yeah. Yeah. Yeah. Totally. Um, and yeah, different people like using, even if it's not a skill, it's just like, remember this for next time you're effectively compounding. It's the same kind of idea, just on a bigger scale. Yeah. Yeah. But, But the plugin is great for the engineering loop. And I don't know, could you use it in a regular workplace setting? Definitely. A lot of non-technical people in Co-Work, for example, use it all the time and love it.

59:54Corey Noles:That's awesome. I'll have to try that. There's something about the models right now, which I think will probably always be the case, where if you just get them to think more or use more tokens on your problem and do more research and spend more time on it, you just get better results. and I think compound engineering the plugin is a hack for that where uh all the things that the model comes back with and you're like I don't know if this is like totally right or like it should have done a little bit more research here it's just really good at getting the models to do the maximum possible amount of work so for important stuff or stuff that requires a lot of thinking it's a really good workflow to use you know I've gotten into using coding agents for everything now like i mean even codex i've now built out codex to do so many things that have nothing to do with like projects i'm working on like it's it's very much you know skills and automations and a little bit of everything and like i've gotten to where i love to write in it because it doesn't use like mdashes and a lot of the normal ai tells are gone which is an interesting thing because and as grant put it code bases won't tolerate that no no that's exactly what i said i'm like no you're not gonna have m dashes in your code base what are you that's really interesting wait so what are you when you write with it what's your workflow uh it's very simple i usually have i have i have a number of different skills that i use in there for like i don't know this is this is my sub stack this is a blog article and you know and i've essentially over time just managed to extract what a Corey blog article is from, like, my favorite pieces I've ever written over the years and had it analyzed them and pull out the qualities that made them me, I guess.

1:01:38And then just build a skill where it's as simple as saying, hey, I want to write this, use this skill. And you like that better than using 5.4 in ChatGPT? Significantly. That's interesting. I really didn't think, I mean, I really thought, you know, the ChatGPT app was something you'd pry out of my cold, dead fingers. but I've really grown to enjoy it in Codex specifically. I think this is a really interesting thing. I mean, I use Codex much more than I use ChatGPT for things that I used to use ChatGPT for. And the thing that happened is... And when around the time when me and Kieran were having this like whole realization about 3.7 and do you need to touch the code and whatever.

1:02:26That was a little bit after that was when GPT-5 came out. And GPT-5 was a very interesting model release because they continued, even though all of this agent decoding stuff was happening, they continued to push forward this split between regular knowledge work and vibe coding happens in chat gpt and like professional pair programming happens in codex yeah and they they stuck with that split from gpt5 until really until like the last two or three months yeah even maybe december even at yeah exactly and that i think that really hampered them because what anthropoc is able to do is cloud code a like people started to be like there's a whole new engineering paradigm that i can't do with codex because it's too hobbled it doesn't let me do this and then people were like but i could also do it for all this other work and so it started to explode and i think chat gpt got left behind a little bit because chat gpt as a as an app it doesn't have access to your computer it's like it has a desktop app but like it's that's has always been sort of a sideshow and it's really like this chat thing like yeah you can use it's just like the chat gpt website just just living in an app is what it was yeah and but you can like double tap to open it exactly and cloud code is just much more from the beginning an agent that you hand things off to and that has access to your whole computer and your whole life and that's a much better much more fertile ground to grow more power and more work it can do more work for you then okay it's a really a website and now it's a mobile app and now it has a desktop app but like it just doesn't it doesn't it didn't work for them and i think that they to their credit even though it took a while they realized this over the last like two or three months and like totally pivoted codex and we're leaning into it super hard and i think one really good piece of evidence for that is the super bowl commercial they ran was for codex it was not for touchpt yeah right um and i and you you hear all this like all these rumors about the internal kind of they're reorganizing they cut Sora whatever I think that they're realizing this and they're like going full bore into it and then I mean so far I really like it was smart I think cutting Sora was smart like you know the fact is if you've got a big model around the corner and it needs GPUs Sora was not cheap yeah like like it needs to somehow be either making money or bringing in new users or something to justify by not redirecting those GPUs over to putting out something really cool.

1:05:06Corey Noles:Yeah, the latest is that they're going to do basically Atlas, ChatGPT, and Codex as one app. I mean, that's interesting. I mean, Codex as an app on its own is just pretty great. It really is. I worry that if they try to combine it all together, it might not be as good as what they have now. I agree. I know that they're going, I mean, just not from any internal knowledge, but I do know just from looking at their tweets that they're really – I think they're really looking at re-architecting the Codex app, but hopefully in a way that's not – we're making a chat-gb-gb-gb, but we've realized how powerful this can be, and we're pushing it as far as we can.

1:05:44And I think they're going to do a good job. They've been working on it in a way that I really like too that reminds me of early perplexity where you'd see Arvin Srinivas on Twitter at night, and he would be like, what do you want? Give me a feature. Give me a feature. and it's that way on twitter right now every single night with the codex team and they're like what do you want well why do you want it how would that work and like asking follow-up questions and then i think uh i think that is such the way to build in 2026 is is like get that feedback right there i mean yeah the fact of the matter is if these are the people you want to please find out what they don't like and uh it's tough though because of course you know you've also got to separate you know signal from noise yeah which is which is tricky but i bet they're pretty good at that it's tricky even with a newsletter you know sometimes like where's like when feedback comes it's like okay is this do we care do we not do we is this an angry person is this a person with a good idea hiding it behind being a jerk uh you know it comes all ways

1:06:47Corey Noles:two two more things because i know we're we're over over the hour mark um that i want to touch on your product suite. We mentioned agent native. Are all of your tools agent native now? How would you rank them? If you could tell us a little bit more about that. The other thing is just any advice for building and shipping in the agent area. What's really funny is if you talk to anyone on the Every Team and you ask them what my favorite word is or phrase, they'll say agent native because I just cannot stop saying it. Well, you coined the term as far as I know. I think so. I think you did. And it's just like a thing that I realized in December.

1:07:27And then we came back from Christmas and New Year's. And I was like, agent native, agent native, agent native. Is your app agent native? Is it? Is it? Is it? And we're really getting there. So Quora, which is our email agent, it has a CLI. And that's really cool. And we're launching a new inbox for it. So you can just manage your entire inbox with Quora. and it has an agent or any one of your agents can use it. Awesome. I think that's going to be amazing. Spiral is our ghostwriter with taste, and that is definitely agent native. That has a CLI. So when I use my Plus One and I'm asking it to write tweets, it'll just go to Spiral, talk to Spiral about what it's trying to write.

1:08:09Spiral has my voice and style, and it just goes back and forth and gives me a few options, and I think the writing is much better.

1:08:15Corey Noles:So the agents are talking to each other. The two different agents in two different platforms are talking. That's wow. And what's really cool is like, okay, so Spiral has this interview mode where it, um, in order to do good writing, you have to download a lot of context from who you're writing with. And you can actually do a lot more context, download agent to agent than human to agent more quickly because my plus one has access to my whole life and Spiral. We would normally have to like build integrations for all this stuff, but we don't have to anymore because it just integrates into plus one and it just, it changes a lot.

1:08:47um yeah so spiral is agent native uh monologue is our speech to text app it's like whisper flow we're basically what we're doing it's not really agent native it's like it's kind of in this like weird category where it might not necessarily even need to be but what we're what we're doing is creating a way where you can whenever you activate it on your phone or your computer whatever you say you can direct it right to your plus one so that uh it doesn't go into the app that you're in but it's sort of like having like a walkie-talkie with your with your plus one that's cool and i think that's going to be sick um and then sparkle is it has we're launching a version soon that has an internal agent so uh i would say like what is sparkle again sparkles are a file organizer the file organizer yeah okay um and then obviously proof our document editor is agent native so i would say we're like 70 ish and we're really getting there and and we are as i said like you have to kind of reset your product suite every three to six months if you really want to take advantage of what the model models are capable of and we're definitely like

1:09:56Corey Noles:in that process right now yeah do you worry about like the models becoming to the point where you know it's not worth it to build your own software i mean i know some people have that concern i take it you're not in that camp i definitely not in that camp because my my answer is just surf the models every time there's a new better model you can build a new better product right on top of it that the model can't do by itself yeah that's and that's our job yeah awesome love it um and then i guess last thing is just building and shipping in the agent era you've learned a lot with proof we've kind of talked about a lot of things but just what is what is your advice to anyone who wants to build their own products with agents to use agents in this era so same same advice is um if you want to make sure absolutely that you have a job and you're thriving in ai surf the models yeah your model comes out use it push it as far as you possibly can and i guarantee you there's no just because of the way llms work there is no way that the that that model is going to be better than you at using itself like it's not trained on itself so you're going to be finding all these different new places to push it and if you just surf the model that's going to be like a really really valuable place to be and um that's what that's what we try to do and i think that's my number one piece of advice for for anyone i think for engineering or stuff specifically is like we're we're in this world where you can build anything and so if you are if you're someone who like has ideas and wants to be shipping stuff and you're not like you have to really question like why um because you could be and so it's probably something there's something there to to work on if you're someone who is building lots of things and not finishing that's also a problem um this stuff is it can be actually kind of addicting and so that's making sure that if you're using it vouch i can vouch for it yeah totally right um if you're using it with a particular goal in mind and you're not hitting that goal it's worth like really assessing that and evaluating whether or not it's doing the things that you want.

1:11:59And I would not be too precious about it. Just get stuff out there as much as you can and see what works. And it's just a really fun time to be building things. It is.

1:12:08Corey Noles:Yeah. Do you have any particular way you like to ship stuff? Do you use a particular platform? Any advice on the technical side? I mostly use Codex now. I use Claude a bit. I think Claude still has better empathy. So if I'm trying to design an API that agents have to use, I'll ask Claude, what's the most ergonomic way to do this? I also think Codex's design skills are a little rigid. And so if I'm trying to do a good UI, Claude is very helpful, especially if I'm not doing it with a designer. Like if it's just me and I'm riffing. um that's my experience with codex too is that if if you are a front-end designer you can probably get it to do exactly what you want but if you're not you're going to get a better result from code on front that's yes exactly um i also i have to say like check out proof um use it with your agent it's pretty sick and if you want more stuff like this check out every every.to um we publish stuff like this all the time and there's always new things coming up.

1:13:20Thank you so much, Dan.

1:13:22Corey Noles:Yeah, we really appreciated this. This was awesome. Great to meet you and talk about all this stuff. Let's see here. I'm just suddenly realizing we don't have a plan beyond that moment.

1:13:37Corey Noles:It's totally fine. All of a sudden I was like, oh, now what? So, I guess what we can do is we can just kind of recap what um what what we just talked about and then um a lot of wild news this week too worth maybe tapping into if we wanted for yeah we can do that yeah let's let's take a look um but first of all i just want to say like you know maybe some people are familiar with um dan's work they've already you know read some of the stuff we talked about um but you know obviously he has a ton of great resources his open claw installation guide is like the best um but it sounds like if you want just like a out of the box like solution for that that works with slack plus one is is the the tool there and then um yeah i just think like his his whole journey of proof i think for me the takeaway point there was you know definitely before you start building a project when you're in the planning stage make sure that you have it go out and research you know or you yourself research the best practices for using the tool like what i'll do is if if um you know codex or cloud code comes to me and says, Hey, here's the plan.

1:14:46Corey Noles:We're going to use XYZ library. I then go and I ask it, why are you choosing this? What are the other alternatives? Um, why is this the best option? And if you don't know, go use web search and, and find out and tell me, but I like the best practices. It's like a good best practices is a good, um, way to kind of condense a simple way to, you know, you might even want to provide those best practices. If you, you know, if there's a site you trust more than others maybe go find what they have on those best practices and and drop that in there could probably save you a lot of headache i much like grant kind of my strategy is i don't want to ship it until i understand what it's doing yeah you know like i want to understand the code at least at a rudimentary level because you know i i don't do this i didn't go to i don't have a degree in computer science or anything i'm just kind of learning as i go so I'll take it and drop it into chat GPT or I'll drop it into even Claude or Gemini.

1:15:44And I think it's, it's really good to be like, just walk me through this. Tell me what we're looking at. What's, what's the architecture here? What, tell me the framework that allows this tool to work. And as I'm saying this, I'm thinking I didn't do that with the one I just built.

1:16:02Corey Noles:Look, it happens. It's, you know, the coding, the coding part is trivial, right? So you can always rewrite the code. I'm always telling people to write better prompts, but the truth is when it's just me left to my own devices, I'm like, do that thing. The two things I wanted to ask Dan, which I didn't find an opening for, is if and how he uses AI and which model in his writing process, because I'm very curious. They're very high taste over at every. I think everything that they publish is really well done. They put a lot of thought into the ideas and the writing, and they've slowly embraced using AI as part of their writing process, but I don't know where exactly they are with that these days.

1:16:44Corey Noles:But everything that they publish is really great. And then, so that was one question I want to ask him is, where is he using AI in the writing process? Sounds like he's doing a lot with his agents and Spiral and all that stuff. Another thing I wanted to ask him, actually, I forgot. Forgot what the other thing was. But there's another thing. If it comes up, I'll message him. We'll come up with it later if it returns from the memory lagoon. What all was there this week that's been crazy? We talked a little about Sora. Yeah. That's really interesting. It's been hilarious to me watching the way these different companies are shipping.

1:17:29Yeah. Because they're all shipping like insane right now. But they're each doing it in different ways. Claude is doing public releases that are a normal thing. It comes with a blog page. Here's a release, and they're going everywhere. With ChatGPT, it's more like a tacked-on feature. Yes, with ChatGPT, it's more of a tacked-on feature. With Gemini right now, I'm watching them daily throw things out for Gemini CLI, for Google AI Studio. They're all doing it. And the funny thing is the part of this that I get such an incredible kick out of is how often it's a thing the other company already had and the people go crazy about it.

1:18:12And that goes every direction. Like what that tells me is that most people have a favorite and never leave it.

1:18:23Corey Noles:Yeah, I think that's true. And then once yours gets the same capability, you're like, yes. This is amazing. This is a great big step. It's like actually the other one said that. It's kind of like when a show goes viral when it hits a certain streaming app. Like not everyone has Peacock. The show is always streaming on Peacock. Like no one knew it existed yet. But then all of a sudden it hits Netflix and it's like a show from 12 years ago and it goes crazy viral just because everyone can watch it now. I was laughed because like computer use. Claude did it first. ChatGPT made it better with ChatGPT Agent.

1:18:53It made a big loop forward with 5.4. And then Claude came back and made another giant leap forward this week. And every one of those four is like computer use never happened before it. It's absolutely like nobody ever heard the term before. And I'm like, that's so funny.

1:19:09Corey Noles:Yeah. No, but look at this. So everything Claude shipped in the last 52 days. This is a calendar. This is everything that they've published since, what's the first day here? February 2nd. And almost every day, a couple Sundays they missed, a Thursday they missed, maybe because like you know some there was like a red alert or something like opening i dropped something i don't know you know a saturday they or a monday they missed but it's just it's just like something is going on over there where they can just build and ship things as quickly as they can um some people think it's like they just have clawed six internally i think you're watching them doing exactly what all three of them are doing because like gemini's shipping like what 80 features a week or something, you know, small ones, but they're all features.

1:20:02They're all shipping so incredibly fast right now that to me, I'm always hesitant to say auto-recursive, but I don't really have a better way to say it than they're really leaning on the models to build themselves.

1:20:20Corey Noles:No, I think it's true. I think they have internal workflows where the agent is writing the code, the humans are reviewing it, the agent is reviewing it, and then they're shipping it. And you should assume every one of these companies is one to two models ahead of what is public right now. Yeah. That's also a fair assumption. Anthropics is not building with 4.6. ChatGPT is not being built with 5.4. you know I mean they're absolutely working or excuse me it may be being built with it but you know they're absolutely working with these models a generation to even two generations ahead depending on the company and I'm sure Google's doing the same thing as you should be you know yeah I'm trying to look at see what's happened since because Thursday is a big day when they publish a lot of stuff let's see what's happening here open claw the iPhone of tokens

1:21:19Corey Noles:interesting well we can look at um we can look at what was published yesterday as well the iphone of tokens is such a weird way to say it like i get what he's meaning he's meaning like it's it's it's tokens having their iphone moment kind of yeah exactly um but yeah we have the round the horn die dash that we publish now where i you know look at twitter every day i look at all my other sources that I checked. You're doing them daily now. I didn't even realize. Yeah, I'm doing it daily because it's just easier for publishing based on... I'm not logging these. I need to be logging these because they're not coming up in the right category.

1:21:53Corey Noles:Yeah, but yesterday you published... Oh, that's not the same. I thought that was the one I published right before we went live. Oh, no. This is Codex 101. Did you publish another one? Yeah, I published Codex 10 Tips for Non-Coders. Oh, cool. Awesome. I wanted normal people stuff. you can do in those tools. I think you have a good cloud code version of that too. You published the 101 guide, right? It's okay, I'm just promoting you. All right, thank you. And I kind of distilled it down to like seven tips here and then I clicked to the full guide. We'll have to plug the other one that you just wrote, which is really cool.

1:22:34Corey Noles:But I really like this. I think this is a really great way of introducing people to Codex. I think the, you know, to Dan's point. It applies to all of those coding agents, I would say. Oh, sorry. Yeah, no, I agree with you. I agree with you. And I also would say, let's switch off of this for a second. I would also say that it applies to, like, using codecs for non-coding tasks as well. Yeah, that's fair. Like, I think, you know, just like Cloud Code was good for not just coders, but everyone. and then they made Cowork, I think Codex is the same. I think it's a good app that's really good for coding, but it's also good for just using it with files on your computer.

1:23:20I think the app is the least intimidating way to use it. I think there are, and I would say that with Claude too, because I just, I feel like less technical people are terrified of a command line interface because normally if they interact with a command line interface, it means something is broken on their computer and this thing is flashed up. So as opposed to, you know, it's unless you're the, you know, the DOS generation. And I like that the app makes it a little more inviting. You know, there's still some coding lingo there, but it really doesn't matter. You could absolutely ignore much of that and just go in and use the automations and skills and pick your models.

1:24:08Corey Noles:so far really impressive but yeah this is a great this is a great resource for people it also pulls from the official codex docs and and other stuff so it's like a great starting point if you've never used these tools before and you want to do what dan did which is build an app and produce it and hopefully you know not not uh skip the best practices and you know not sleep for seven days but yeah but seven days of stress and sheer terror yeah yeah um uh but uh but Yeah, it's a great resource that you published. What else happened yesterday? Yesterday. The ARC Prize. We were, oh, this was the other thing I wanted to ask Dan about was benchmarks because you and I were having a discussion right before we got on here about the value of benchmarks and whether or not they're useful.

1:24:56Corey Noles:And we can save everyone the back and forth between you and I, but I think your ultimate conclusion makes sense to me, which is benchmarks are toast, like they're cooked. and the only benchmark that really matters is a list of tasks that the ai can accomplish and you just list every human task you can think of and just mark off and i don't care what you connect it to to get there i don't care if you're doing that in cloud code i don't care if you're doing it through an api but i think if there's a thing it can accomplish that is really good we should see that uh and and honestly i think a lot of research labs don't know what those tasks are in some cases until it gets out and it's in the hands of a really gigantic diverse set of people um but like i i would love to see like i need to see it like you know if you're going to test things like creativity for example i don't want to know a 46 versus a 51 i want to see what it wrote versus what it wrote and a lot of these are very closed right you know i i think because what's going to affect me and not just me because like you know i i tend to think of you and i as normal people and then i remember that most people don't read as much of this as you and i do the fact is we're probably a little married and i feel like i'm behind but then i compare what i know to the average person and oh my goodness people are not prepared the gap is getting bigger and it's getting big faster.

1:26:28And since I'm going to go back and say Gemini 2.5 Pro. From Gemini 2.5 Pro forward there have been these big hops from model to model, from tool to tool as they've come, and the hop keeps getting just a little bigger and a little bigger. And I just don't see a I don't know. I don't know how you catch up today

1:27:00Corey Noles:unless you're stopped by the Neuron.ai and subscribe to our newsletter and enjoy our fine podcast if you do those things you are sure to be well informed so for anyone who's watching I don't even know how many people are still on the stream at this point but for anyone who's watching this is the ARK AGI 3 and basically it's a series of video games that test your ability to reason and adapt to a new situation, figure out the rules of the situation, and then basically try to solve this puzzle. And I have failed this one three times because I missed a key thing there. Oops. Now I got it, and I'm going to try.

1:27:44Corey Noles:This is kind of hard. Yeah, go ahead. We're only on LinkedIn and X. apparently around 12 o 'clock we lost youtube okay well that's it is what it is i mean that's why the chat's so quiet yeah that makes sense yeah because i was like really nobody has any feedback no that makes sense um so we can wrap this up here in a minute how about at the half hour mark i think so i think that's good i think it's good this is game looks fun though great yeah this is arc agi test so um it's meant to test how well an agent can adapt to a new situation um and it's uh we we wrote about this in today's neuron but basically what what uh what happened was every you know frontier model tried this and was at like less than one percent ability to complete agents specifically models they've stripped all the models and everything else yeah because they're trying to test the actual underlying model and how good it is and perhaps that's not really a fair assessment if the way that we're actually going to be using these things in real life um is with a harness and as a part of a system so i sort of agree with uh cory's point there if you want to expand on that you can yeah you know uh part of my thing is that i mean i do think they're good metrics to know but i don't know that any value comes from knowing this one's a three and this one's a 13 other than yay my team's winning like i feel like that's what we get out of it and uh what i think is necessary is a much more practical approach i think instead of you know playing benchmark whack-a-mole maybe we could move into more of a you know task specific stuff and i don't mean your numbers go on tasks i mean show us examples show us you know maybe it's uh you know here's a 200 page, you know, research PDF that gets shared out.

1:29:45That's showing us like, all right, now here's what their last model did, or here's how I actually think we're so past.

1:29:50Corey Noles:We're so past the point. Like, like I'm going to give open AI a bit of grief tomorrow about their ads. I'm just kidding. I'm going to give it over to a grief tomorrow because their ads and chat GBT are so generic. I'm like, we have generative AI. Like you could make generative, you, you know, of UI at this point, and you're going to give us a little tiny image and ad, like, I get it. Like, we don't want the ads to be obtrusive, but at the same time, like we're past the point where you should be putting out PDF documents. Like, like you should be able to build an entire website with videos embedded with all of the tasks, like showing them, like, come on, like we're, we're way past the point of like PDF research reports.

1:30:33Corey Noles:Like you got to build entire websites for this stuff. Yeah, I want living research papers that I can ask questions to when I'm unclear. I want all of the things. Yeah. There was a great post I saw yesterday. I think I included it in Round the Horn Digest, but it's someone who's saying there's two different types of websites that will exist in the future, and I agree with this. The first one is websites that make it really, really easy for agents to read them, like super easy, like basically like they're designed for agents. That's where your element is. is coming from. Yes. The second is, um, designing for humans and making them as visually stimulating and as interesting and as complicated as possible.

1:31:17Corey Noles:So in one world, you make the website as simple as possible. And in another world, you make it as complicated and dynamic as possible. And I think that's where generative UI comes in. I think that's where you're going to have websites that feel dynamic and alive and like you're playing a video game, but you're on a website like like i just think that like that's where it needs to go because we have the tools to do that stuff so much easier now so like now the level of complexity needs to go up and and really just like meet people where they are like yeah if i'm gonna read your website you know make it interesting make it cool i can't stress enough that meet people where they are is important and and i think that's what's what's soured me a little on benchmarks uh you know and I think it's important that we begin to try to make this stuff more accessible and explain to normal people what it means.

1:32:06I think it's important that more people than ever are unfortunately picking their sides in a battle as opposed to trying to figure out what this means. It's becoming very political, which I hate.

1:32:22Corey Noles:I think it needs to happen, personally. going political? Yeah, I do. Because I think at this point there needs to be some sort of backlash to the progress. I think the progress is happening so quickly that if there's not a pendulum swing the other way then who knows what happens. And I think the natural order of things is for the pendulum to swing. So I think that's healthy. I think it's probably healthy. I'm not saying I agree with the form that that backlash will take. Let me put it that way. At what cost is my concern? Like, I mean, you know, and their argument would be at what cost are we going forward?

1:33:02And mine is, I just have a genuine feeling that whatever country leads the way in this is going to be leading the world for a long time. I think that's fair. And that's the thing that has made me, I won't say anti-regulation. I do believe there are needs for regulation, but I would rather see them around things like deep fakes and child safety as opposed to pausing and preventing. I think coming around and kind of cleaning up the mess, perhaps you could legislate something like, I don't know, affordable training methods, ways people can learn and re-skill quickly, new ways to start revisiting income and what that's going to mean in a later world.

1:33:49I think like I'm absolutely for certain types of regulation. I just – I have this fear that the only two options are foot on the floor or slam the brakes because that's how we do everything in America now is everybody's extreme. It's like everybody slow down. Let's live in the gray a little here. Let's find a medium.

1:34:12Corey Noles:Well, that's why I want the pendulum to swing is so that it comes back to the middle. Hopefully come back to the middle. Yeah. Maybe it'll look like a pendulum over a skyscraper where it takes out a couple rungs on a rope. If you think of it like a pendulum over a rope in a Rue Goldberg machine where it's slowly cutting away at the rope as it goes lower and lower. Yeah, yeah. Oh, what movie was that? That takes me very Edgar Allan Poe. Yeah, exactly. I know exactly. Mergers of the Rue Morgue, maybe? Darkest Sour? I forget what it is. But I remember the guy laying on the floor and the blade flying over.

1:34:56Corey Noles:It's just getting closer and closer. Yeah. Yeah, I think also, I think this capability unlock of where we're going next is going to necessitate we redesign our society in some ways. I think it just has to. Yeah. And so it'll be interesting what form that takes. It's hopefully the form is not centralizing power. Hopefully it's decentralizing power. As divided as we are right now as a country, as a human, maybe that fresh start's what we need. I don't know. But I would listen to all of you. I think it's just going to necessitate it, yeah. It's going to lead to – and there's some interesting ideas.

1:35:37Corey Noles:There was a great Axios article yesterday that I'm going to basically include in tomorrow's newsletter about um centurini research published some ideas for um i think gina raymundo who's in the um who's in the um government like the department of commerce or something she had some ideas um and there's a lot of ideas for like ways that you can build out the safety net um to incentivize using humans and keeping you know people relevant to the workforce you know as these tools and capabilities increase um so there's some really cool ideas around that that that don't look like put on the brakes do nothing yeah or or bernie sanders is like pause all data center i love that he goes and talks to haiku out loud uh oh yeah that was funny that was funny it was the the best meme of that is old man yells at claude i was like no you send him to the one that's trained to be unsure about its consciousness that's terrible that is not the one you you well it's just funny because in the video you can see him totally getting claude pills where his like whole tone of voice changes he's like oh somewhat reasonable like i feel like he was prepared to fight the ai or something and then yeah he actually found it you know i actually like you better than most humans yeah yeah well like wow you i asked you a question you answered me honestly what without yelling nobody's screaming no names yeah oh goodness anything else gory I think that's it I think it's it for today everyone thank you so much for joining us if you haven't yet please remember pop in the chat go find a link to the giveaway you got your last shot to get in and get a chance at a DGX Spark and you should because it's pretty sick also make sure you check out the video we dropped last night with Nick Heiner from Surge it's really cool we dropped three different ones last week Nicole Bayer from Carta we dropped Dr.

1:37:33Chi Chow Hu from SESAI we dropped Kerry Briskey from NVIDIA who is an amazing person Eamon McGuire

1:37:40Corey Noles:from Proton yeah yes Eamon McGuire from Proton yeah you know lots of these you should go watch some there's great stuff and we appreciate you being here we appreciate your continued patronage to our fine publication and I don't know I'm just being a nerd I love it have a great week everyone and we'll see you back next time farewell for now humans