Why Anthropic Thinks AI Should Have Its Own Computer — Felix Rieseberg of Claude Cowork & Claude Code Desktop

[MUSIC] Hey everyone, welcome to the Latent Space Parkest. Our first one in the new studio, we are home. This is Alasio from the Recurnal Labs and I'm joined by Swix, editor of Latent Space. Yeah, so nice to be here. Thanks to TJ, Alasio, Alan, helping to set everything up. It looks beautiful. Maybe you have the logo outside. Yeah, it's got a thing. It's like really nice, right? When you walk in here as a guest, you're like, "Oh, this is a serious production." You like feel it immediately? Yeah. Felix, you've been your currently product manager of Corek or. I really reached. I actually, the identities are kind of a member of technical staff. I know. Remember, I think that stuff is like the official title, we'll carry around forever. Yeah. I recently kind of wanted, like, we've been obsessed. I've been using it a lot, even for managing Latent Space. Corek helps me upload videos and like, title things and like, edits and everything. It's like really amazing. Cool. He's a multiple times Corek is a GI in the group. Yeah. Yeah. So we have a second channel for the INSPACE TV. Basically, this is our Discord meetup. We have like, Cloud Corek is my PEGI. I don't know if we have uploaded it yet, but one of the sessions was like a Cloud Core thing. I would love to see it. I'm so curious. Like, one of the most fun parts of my job is that I constantly see the weird things people use Corek for. Because it's obviously like very hard for us to actually design for specific use cases. We do. But like, every single person who's like most of MACE, is usually amazed about the thing that I didn't even expect Corek would be good at. We have a new designer. And it's one of the first small tasks. I was like, hey, we need like a new emoji for Corek for our internal stack. It's like a pretty small thing. I was like, can you please do it? And he drew an SVG and just gave it to Corek. I was like, can you animate this emoji? And now it has like this beautiful loopy animation. And I mean, I think obviously this goes down to like a ton of such. You can do more things with code than you expected. But it's like that kind of stuff that is really fun to me. So long story short, I would love to see like the kind of things you're doing. I'll pull it up. Yeah. Yeah. But before we get into it, I think always want to start with like the top level. What is Cloud Corek for people who haven't heard of it, haven't tried it out? Okay. Real quick, Cloud Corek is a user-friendly version of Cloud Corek. So the way it basically works is we have Cloud Code and for us fairly impressive agent harness that over December, we noticed more and more people are using either even though they're not technical. They're not at home in the terminal. Or they are at home in the terminal, but they started using Cloud Code for non-coding workloads. I'd like managing expenses or like filling out receipts or organizing knowledge based. Like there was a big obsidian moment that a lot of people liked. And we wanted to capitalize on that, but also bring this capability to people who are not terminal native and who might not know how to brew install something. So Cloud Corek is Cloud Code running in a virtual machine with a little bit of padding, a little bit more Godrails, making it a little safer and a little bit more convenient for people who don't want to first open up the terminal when they go to work. It's interesting that it's kind of pitched that way as a more user-friendly thing. Because I always feel like to me, I treat it as why I'm familiar with Cloud Code. We did a Cloud Code episode a year ago. But this one is even more power-user tools because it integrates much better with cloud and chrome and all the other tooling. But maybe that's a perception thing. No, honestly, I don't think you're wrong. This is a thing I've been thinking a lot about for the last two weeks. But when they say user-friendly, it's like, oh, it's the dumb-down version. But no, actually, this is the superset. Yeah. I think a similar thing happened to me about 10 years ago, maybe 12 years ago, when I was on Microsoft. And we started working on Electron and browser-based technologies and cross-platform stuff. And one of the first use cases was Visual Studio Code, which used to be a website. And the initial narrative was, Visual Studio Code is like a more user-friendly version of Visual Studio. But in a similar vein, I think there were some voices saying, oh, this is not for serious developers. Like we're not going to use this, right? For like anything. And I think in the end, what happened is people have different stories about why Visual Studio Code became such a big thing. But my personal, my personal belief is that the hackability and the extendability is like, played a pretty big role. You can hook in Visual Studio Code that are almost any workload. It's so easy to hack on. It's so easy to build extensions for it. And I think call work might be hitting a similar thing where it's very easy to extend. And it's very easy to bring into your workflows. So the convenience, I think, is a bit of a, it's obviously the thing we strive for as developers. But I think the way people find value in it then is by probably mapping it onto whatever they actually have to do in their job. So, end of last year, you see the spike of non-sacnical usage in Cloud Code. What's the design process to say we should make Cloud Code work? Because I mean, you built that in only 10 days. I'm sure there was some discussion before on what it's easier to use. Mean, you know, like making like a desktop GUI is obviously one way to do it. But like, there's a lot of nuance in the product. Like, maybe talk people through what was like the trigger of like, we should build a separate thing. We should not build like a different Cloud Code thing. And then maybe some of the more interesting design decisions that maybe you didn't take. Yeah. I think, philanthropic, we've been thinking about ways to move people who are comfortable with using Cloud Answer questions and bring more of the power of like this thing to now like execute tasks for you. I can like solve problems for you. I can like build things for you. How do we bring that capability to people who are currently mostly comfortable with like a question answer paradigm within the chat. And we've had a lot of prototypes around that. This is going back as far as like easily a year and a half. Like, we had a lot of people working on that. And internally, endthropic is a very prototype demo first culture. We have a lot of like internal prototypes that don't reach the public. And what co-work actually became is like we sort of picked the right pieces out of the many prototypes that we had. Right. And that's that's maybe also like I think an important qualifier whenever people mentioned this like 10 day number. I do think it's important to me to mention that we didn't stop the scratch. There was like a lot of stuff already happening, right? And I think it's important for people to remember that when you build a website, you use React, you use like a bunch of other things. And this is like a similar scenario with like a lot of pieces where we had. In terms of decision paths, I think we live in like an interesting new world where execution is actually quite cheap. So maybe maybe what you would do that's so crazy to hear. That's why you should be ideas of cheap execution is the hard part. No, I like the we used to live in this world maybe where you would take a product manager and the product manager would go to a number of potential customers. And it's a very low bend with way we would try to try to like tease out what are the problems they're having, what are the willing to buy. And then maybe what can you build to like a dry out that need. And then you go back and you like draft a spec and you think about it and then like you make a design and you execute it. We internally and then throw up pick up not pretty much closer to the point where like don't even write a memory just like build like let's build all the candidates very quickly. So let's just build all of them and then pick the best ones. I think the decision that is most impactful both for the product as well for the users right now is like the way we put value on your local computer. I think that's a big decision point. A lot of people have thought about should this thing whatever it is to ultimately run into computer or should run in the cloud because they're big trade-offs right. I guess like if we solve it off it will be easy to do in the cloud but I think like the fact that I can just download any file from anywhere and then you put it in co-work there. It's like a big unlock. I mean it's interesting you mentioned reusing certain pieces. I think this is something I've been thinking about even with a cloud code right the price of like writing code is going to zero blah blah blah but it actually seems like the value of having some sort of platform substrate is like increasing because as you build this new things you can kind of plug them together. Yeah. So I almost feel like when people are saying oh the value of a lot of software is going to zero because you can recreate it to mean some was like the opposite is like having an existing platform to build on top of it's like even more valuable because you can kind of bolt things on. Yeah. You have obviously MCPs you have skills you have like obviously the models which is a big part. All these things kind of come together. Do you feel like that's a valid way to think about it where people should invest even more in kind of like these primitives to rebuild on or are you like recreating a lot of it each time because like things change and it's easier to rewrite than we use. You know I think you're right. I think you're right that the holistic platform is really useful and this is maybe a whole like somewhat contrarian view to a lot of people in AI. I actually don't think that the future is going to be hyper personalized software down to the point where everyone is running the own version. Like I actually think it's going to be quite hard for one of us to have our own internal chat tool and like if I want to talk to you like how are you going to work right. In the in the kind of stuff called again how we build it I think it's a bit of a combination like what the execution that gets cheap is not necessarily rebuilding all the primitives. I think our priority there's also not a lot of value in it. So for instance my team did not think about rebuilding cloud code we're like very much started with the with the core thesis of this should be cloud code and then we'll like build things on top of it. The part of the execution that gets a little cheaper is like how do you take all of these Lego pieces and put them together in a way that makes sense for users. It is actually valuable. You have so many different approaches now in terms of what kind of things do you actually elevate to a primitive. Do you strongly believe that all your products should be built by just combining primitives with about it also as a variable do you keep some things on total. I think that's still evolving but I think it was probably going to go ways like I'm not sure if it's going to fully go away, but I'm going to say, I think for me personally, I will probably no longer try to come up with a really good product without testing out with people. This is not a new concept, but wherever you used to have to make costly decisions around to be picked technology, AI or technology B, or do we like build it this way, build it the other way. I'm really strongly believe not, you just build all of them and try them out with that small focus group and then whatever is better is where you go with. And that is probably quite different even from how we maybe worked a year ago. I think this happened very recently. Yeah, I started building something on Electron since you're here, coincidence. But then Electron and SQLite are like, there's some issues that like, be Zoom development and like, building, anyway. And I was like, let's just rebuild the whole thing and just re-graded the whole thing and it's like, it's done. Yeah, no, it doesn't take any effort. I don't even know Swift. Yeah, exactly. I was like, I'm not reviewing it anyway, whatever you're writing, whatever language. You pick. But the important stuff that I did was not write the Electron bindings. Yeah. It was like the logic of what happens in the app. And then the model is like, yeah, I can just recreate the same thing as with. Yeah, I think you still want, especially for people who are doing like high performance software, like a very complex software, you still want like to some view of the architecture. But you can use the mark down for that. Right. Yeah. You don't actually have to read the code. Again, I'm still like on a, sort of like a definitional thing. Can we build a good mental model of Cloud Core work? This is what I have, right? Like you said, it's like fundamentally Cloud Core, we don't want to touch it. There's the Cloud app. There's Cloud and Chrome. I think you guys do something different in planning. But I've been talking with Tarek who is on the Cloud Core team and you guys are like, no, we just exposed planning. Maybe you can clarify what are the major pieces that people should be aware it goes into the core. Like, okay, I think you basically have them. So you can take planning more or less out. I think that's a few things that are really valuable in Core. The virtual machine is probably the most powerful thing. So we currently run like, we currently run like a lightweight VM and we put Cloud Cloud into the VM and we do that for a number of reasons. Safety and security is a big one. But even if you, even if you ignore for a second safety and security and you're just like, okay, you'll all, I want this thing to do whatever. It is quite powerful to give Cloud as a computer. That is like generally a good idea. And in terms of architecture and UX and everything else that we've been working on in a through a pick, it often is quite useful for you to like, and from authorize Cloud aggressively and just be like, this is a person. Yeah. What would you do if you give them if you had a person, right? And the analogy I've given my dad this morning, who is still like quite insistent on using chat even for like coding things is if you are a developer and you employer told you that you don't need a computer, they're just going to like send you emails with a code and you send emails with code back. Like that may be worth a petriot loss in the back, but that is not very effective. So what we can do with the VM is because it's a Linux system, Cloud code has more or less free range to install whatever needs to install. You can just love Python, you can install Node.js. We do have strict network ingress and egress controls so you can still as a user in like plain human language make it clear to the entire system what you're okay with and what you're not okay with, but in no point do we have to ask a real person like a person who might be in marketing or a lawyer, I don't have to go to lawyer and be like, are you okay with me installing home brew? Yeah. Right? Because the implications of the question and the answer are complex and nuanced and like not not easy to reason about. And this gives us a lot of distraction that makes Cloud very powerful. Now then around it, we do probably have a number of things that also keeps growing almost every single week that you probably noticing that make code like maybe better for certain tasks than just cloud code on a zone. Yeah. But most of those actually live in the system prompt. They're about like, what can we infer about the work that you do? What can we, what can we do? Is into the system prompt to make that more effective? It's of course the like very tight integration with cloud and Chrome. You're noticing that a lot of people, especially as the models get better, a lot of people throw up their hands when it comes to ncp connectors in this era. I'm not going to go through like 25 mcp connectors, click off everywhere and then like half of them don't let me do these things anyway. So cloud and Chrome is quite powerful because we can just talk to the cloud and Chrome start agent and they'll just do things for you. Yeah. So one example, right? In mcp, I honestly, I think the state of mcp is kind of like really hard to integrate. I needed to add a Figma mcp to the coding agent that I use. And I didn't want to read the docs. So I just had called to it. And it's it's greater reading docs. And the same same way, I had to set up like a Google Cloud account for some project that was working on and get some API keys somewhere. And Google Cloud is famously super hard to navigate. So I just didn't want to do with any of it. I just used cloud code. Within the first week of developing on cloud with the 7 very, very quickly, I caught myself by sending you this code for coding tasks, which is not ostensibly what we built it for. Right? I found myself, I found myself like on our internal internal tool that we have for to collect crashes and just like debugging information. And I found myself sort of picking out the ones that I think we can easily fix versus the ones that might be like kernel corruption or something else on the operating system. And I found myself sort of picking these out and then just telling cloud go fix this bug. And I was like, what am I doing here? Go one level up, tell a coworker. I want you to go to all these crash tools. I want you to find all the bugs that you think are fixable and not like an operating system crash. And then I want you to tell another cloud to like fix all of that. And that's that's that's what I call another cloud. Yeah. So it can spin up another instance or it currently what I do is and this is a bit of a hack, but I tell it to use cloud code remote to just call it itself. Yeah. That's interesting. So you basically take if you imagine like a dashboard with like 20 bucks, you, this is remote control or clock remote. I was sorry, I just wanted to confirm what the way I'm using it is I've got a lot of co-work running and I'm telling co work here's where I normally go every morning to find the latest bugs. Go read the entire back list, separate out which ones are fixable, which ones are fixable. And then for the fixable ones, four is this almost loop for each bug right in markdown file with a prompt. And then for each markdown file that is a prompt start of a closet. So natively cloud code has this cost of sub agents. And this is basically a sub agent, but you're not using the sub agents functionality. And the reason I'm not is because I'm firing that off as a cloud code remote task. Yes, kind of nice because then it can just fire it off. I can go to my next meeting and in cloud code remote now the work is happening. Yeah, you see like you're already starting to use the cloud over your local machine. And I think this is one of those things where like Washington just everything just be called first, right? This is such a good route. I like solely about others. I have so many thoughts about it. OK, so I generally believe that Silicon Valley overall is undervaluing the local computer. And my default argument for that is always how come we're using my books and not like an iPad or a Chromebook. There's like still value in having a local machine. And now when I think about cloud is the entity that is supposed to be very useful to you, I get tremendously useful to you. I think that entity needs to have access to all the same tools you have access to. Otherwise, it's going to be hamstrung like all these complex ways. And there's sort of two approaches we could take. You could say, OK, we're going to like one by one chip away at everything that is a dual computer and move it into the cloud. That's one way to do it. And I think other products have taken their path. I personally, this is a very personal opinion, but I personally, for the amount of tools that I use, just don't have the patience to give another tool like permissions to every single thing and keep those permissions up to date. The second thing that I'm still grappling with and I don't have a good answer for anyone to say it, but the second thing I'm still grappling with is, what does it look like for someone to slurp up your entire work and put that in the cloud? Like if I just, that's an example, like if you could click a button and I just clone your entire computer into the cloud, is that something that you would want? I'm not totally convinced that everyone will. And that is sort of like upstream of all the technical issues we're going to have. Because like in general, I think the world is not ready for this kind of stuff. Like I'll give you one quick example that would probably be very easy for us. So as a desktop app, we in theory with your permission can do a lot of things on your computer, including reading your Chrome cookies if we really want to do it, right? We could take your Chrome cookies. You would have to decrypt them for us, but we could put those on the cloud if we really felt like it. Pretty easy solution that would be super cool because it's like, oh, we can do all the tasks in the cloud now. A lot of websites, thanks included. If they see the same authentication from a two different locations, we'll just lock down your account and now you have to go to the branch and be like, okay, I'm here with my passport. Yeah, actually, no, no, no. Wow. As tired as well, I'm off the term agent for the agentic future. I think there's a lot of stuff that sort of slowly needs to catch up. And until that's the case, the way I as someone who's working on cloud can make cloud most effective is to like put it where you're working. Anything else I thought with our mental model. So basically, like part of me also just what like the more I understand how it works, the more I can use it to its full potential, right? Yeah. And so what I'm hearing from you is you told me to delete the planning thing. You're not doing anything special on the, that's only exclusive to clockwork. We have some tricks for this sort of like change freak over week. We evil, co work maybe against different use cases than evil clock code. Right. Obviously think about it this way. Okay. So like clock code is like quite optimized for coding. And we mostly evaluate whether or not be getting better or worse depending on how good it is at a typical sweet job. And cloud call work on the other hand. valued more against typical knowledge for the kind of stuff you would find in finance or maybe like a legal office. My personal use case is always managing my, things like managing my personal mortgage or something like that, right? Or like, "Wolf planning for me and my family." Those kinds of use cases, we e-vail, claw, co-work on. And what you might be picking up on is like the subtle changes we make to the system prompt, what we put in the system prompt, how we steer a claw with the tools we give it. It's like either it would be better in one of the other directions, whether there's a trade-off, trade-off success, a lot. Clawed code will be better for code and cloud co-work will be better for non-coding tasks. Will those gaps still exist in the next regenerations of models? It's like a little unclear to me, though. Because right now these hyperoptimizations we make, I'm not sure for how long this goes in many relevant. I think what I was referring to was also, it just qualitatively felt different when I, probably it's just all prompting and reading too much into it. But like the fact that it comes out like a nine-step plan, I can edit the plan and give feedback and see it execute the plan. Yeah, it felt more long range than in cloud code, but maybe that already existed in cloud code and you just build a nice UI for it. It's kind of both. Like if the cloud code people who build the plan in fashion, that is what city they would say, yes, we have one of those things in cloud code and they do. I think people tend to give co-work tasks that are maybe of longer time horizon. I thought he's so long. Yeah. That's like one thing, right? You're just like that the chunk of work tends to be maybe a little bigger. And the second thing is that because the work when it gets longer gets a little bit more ambiguous, we do tell co-work to make heavy use of the planning tool or to make heavy use of the ask user question tool, right? We do want it to come up with like different scenarios of KFT software, the user actually wants. Don't go off to work for like four hours and then come back with the wrong thing and you're probably picking up on that. Yeah. I wish I could tell you and I built this magical thing and it's like there's some secret sauce. I'm born. No, no, no. I mean, it's just tired. It is good. The engineers just want to know that they can plan around it. And I think also for me, I think I have to switch to my other machine because this is a new machine that doesn't have my session. But yeah, the planning is really important for me to like approve or to see whether it's like it's right. And the ask user question is so beautifully presented. I mean, it's also available in like cursor and in quad code. But like, I think like it's so nice to see that it's like it's kind of for me like to understand that it gets me, it gets what I want to do. Yeah, yeah. Just on the topic of eVals, when you say eVal, I think people are very vague about what it means. Is it just like vibe testing or do you have like automated programmatic eVals of cloud core? When we say eVal, what we really mean is that we essentially take the entire transcript, including all the tools that cloud is available ultimately to it. And we then measure what are the outputs depending on what we tweak. Right. So we do run that a lot. We use that in training. We use that in like if you sort of separate out post training from like the scaffolding around it, co-work sort of exists in the scaffolding space. But obviously, we also train on it a little bit. And so when we say eVal, we mean given the certain transcript, what do the outputs look like, including the file outputs as well as the actual token outputs, like the ones that you see in the tat window? I'm curious how much of the failure modes are the model intelligence versus like the usage of the end tool to put the intelligence in. Like, the wall planning is like a good example. Right. It's like one thing is to come up with a plan. The other thing is like make it nice spreadsheet. Yeah. That kind of runs through the plan. Like, how have you seen that evolve? The thing that I grapple with a lot is that whatever scaffolding you come up with, I think we still have a bit of sort of like model overhang with a model is dramatically more capable than right users and for using it for anything part of that is never just not getting the model all the tools to do all the things that's dearly capable of. Right. It's like one thing. And however, whenever you do both the scaffolding, that sort of wondering about point of what point will that scaffolding go away and like how much you invest in figuring out what the right scaffolding is, it's kind of up to it's a little bit of a bet. Right. And one thing that I as an engineer quite enjoys that like working in the throttling and working in a frontier lab, I maybe have a little bit more insight into what's coming coming down the tune in terms of like, what's the next model? But it's the model capable of what is good at what is about that. And I'm increasingly wondering is the right thing for us to like really invest too much in sort of these scaffolding corrections where the model might otherwise not misbehave or just not do the thing that you want. Yeah. Or is it to just give it as many capabilities as possible, try to make those safe so that the worst case scenario is like not as bad as it might be otherwise. And then just simply wait a second for the next model drop. I'm personally currently more leaning into the latter. I think we're going to see a lot of like applications and companies that do very impressive things with AI that in the short term might seem very effective because they're very specialized to individual use cases. But I think once models get better at generalization and get better at like those specific use cases without being super guided on those. I'm not sure how long that's going to stick around. And you can kind of kind of already see this in like skills and NTP servers, right? We've already seen sort of this like slow shift from MCP service to skills. And like maybe good example is very whom made skills. He was initially hacking on something that honestly looked a lot looked a lot like what Koeh does today. It was not a thing like what if Koeh work but for like people who don't want to build code. And he too did that as a prototype inside the desktop app. One of the first use cases we thought of were okay, what are like coding like use cases that could really benefit from graphical interfaces and like from being a little separated from the excellent line code. And everyone comes up with the same answers data analysis. Sorry. I was like how many users do we have today? I'm going to like it's always data analysis. And I think the thing that ultimately led to skills is that we want to connect this little prototype to our data warehouse. And the team very quickly discovered that like instead of building a custom tool for the thing to talk to our data warehouse, they just like made a markdown for them like, dear Claude, if you want to get data, here's the endpoint, here's what the API looks like, you figure it out. And then they're like hand over control. Yeah, yeah, also just like maybe go one step up in the layer of abstractions. Right? Just yeah. Instead of instead of telling the thing, here's the CLI, please call this CLI or here's an MCP, please call this interface shape. But it's like, this is the endpoint. If you want to know something, if you post here, maybe you can do post SQL, it's going to be okay. And net end up being so effective that they started trying the same pattern of like just giving the model a markdown file that describes whatever it needs to do. That the whole thing eventually became skills and we're like, we should package the Zot. This is a good idea. Yeah, we've had Barry Yamaha on our conference and he's definitely got a good idea there. Yeah, I wanted to show you how I've been using Cloud Core work. So this is my favorite part. So this is like me. This is how we run the Discord. We literally at first, I didn't trust Cloud Core. This is my very first usage. Okay. Right? So then I was like, okay, I will just try to manually download from Zoom all my recordings and upload it to YouTube because this is a very laborious process. I got to click, click, click YouTube isn't super user-friendly. And it just did it. And then I was like, actually, you know, even the download from Zoom part, I should also put into Cloud Core again and I did it. Right? Here's a bunch of it and it starts compacting here. And it starts to even be able to do things like look through the individual frames of the video to name the video so that I can upload it automatically. And this replaces my job as a YouTube player. We will forever appreciate your creative. Yes. And so that's great. But then by the way, it compacts and makes it makes like a new thing. Right? So I don't have the initial thing. But then I asked it to make its own skills so that something that's repetitive and one off and human guided becomes more automated and I can use the skills independently and reuse them. And then obviously you can write skills and that goes into context and skills at the bottom here, which is which is so nice. So I have all these skills that are now sort of you on a weekly basis. I know you've released scheduled core works, which I haven't done yet. But it was really to closely try them. I think this is like so wonderful and fun for me to see because I think one thing that is very fun for me about skills and particulars that they're so easy to make like anyone can make a skill like a text message copy skill and they can be so hyper personalized to you. And this is like sort of the subtraction layer, right? Like I'm just guessing but I asked you a heck. You're very good at your job. You've already given this thing some guidance about how to do it. Right? I just said wrap everything up into into a skill. Right? Yeah. And then I was like actually sometimes I might need to break things apart because some parts fail or some parts might be needed individually. So I pulled it to split one skill into three skills. So it's like a skill splitting thing and then there's like a parent skill that just orchestrates all of them if I want to use that. I think that's like really good. There's one more part, which is the Google Chrome thing that I told you about where I'm like, okay, you know what's better than uploading using cloud core work to YouTube? Like actually looking at the docs to like programmatically upload to YouTube and then putting that in a skill. And I've never done that before. I don't want to deal with Google Cloud. So fuck, or does it for me? That is like so. So so I just I don't care. I just like do I do a thing? I don't that doesn't really matter. That is really cool. And then you I assume paired the skill just with the center if it's built. Yeah. And then I just update the skill. Oh, that is beautiful. Yeah, that's wonderful. It's kind of like a skill. Basically, I think like the way that people ease into cloud core work is like take a knowledge work task that you would normally be clicking around for and then try to turn turn that and then you do the, okay, well, what if you went further? Okay. And then what if you went further? What if you sort of expand the scope of core work as you gain trust with it and also teach it how to replace you? Yeah. It's like a little bit like playing factorial but for your own life. Like you say you start really small. Yeah. You start automating something really tiny. And like once it clicks, you keep adding onto this like automation empire, just like make your life easier and easier. My favorite skill has been every single morning, core work starts looking like calendar and make sure that there's a conflict because people tend to get a lot of meetings. Sometimes last minute, sometimes miss it, soft and painful. And a lot of products have existed like that a lot. I've written in the custom prompt there. I haven't made it a skill, honestly, should. Yeah. But I've given it like pretty clear instructions about, okay, here are some people. If they book over other meetings, I'm probably going to go to their meeting. Like if Dario schedules a meeting. Not try to reschedule Dario, right? And I think there's some other rules in there about like what kind of meetings I care more about, what kind of meetings I care less about. What is okay to like maybe punt like when I want to be one of them want to be working, when I don't want to be working. And it's those really small things that I can think and I click with people right when we launch core work. I think one of the usually says that when most viral on Twitter X was clean up your desktop, which is of course silly. That's such a story. Like you don't need to model the clean of your desktop. Not really. Like this, clean up my desktop. Yeah, exactly. Yeah. I need to choose my desktop right. I guess. Give it access to my desktop. Yeah. Okay. Okay. This is very scary. We'll do it. I did it with my downloads folder. It was like you have so many term sheets and there's like eight copies of your rental lease for your office. I was like, all right. Like don't yell at me. It's like it's such a small task. And then like I would never go out there normally otherwise until people I've built a product that can organize your photo. Right. Because it feels small. But I think to be a point like. Here's just the here's the ask you the questions. Yeah. Beautiful, right? Lead obvious junk. You've partitioned click that. No. If he's not direct. It's not this reversible. I don't make up plan. Yeah. Yeah. You know, I have a typical everything is super messy folder. So yes, I think this is super helpful. So this is a pretty simple task. But I've okay here it is. Here's the progress. I don't see this in this one. Like this got to be something different than than cloud code because I'm like we do the add us. We do system problems that we're like, all right. We want you to think about like this task. Yeah. And then I can I can do like little suggestions for for this things. It's beautiful. Look at this. I can I can like say like, oh, don't do that. Don't do this. It's amazing. And so happy you like it. I mean, the other way around like we're part of the cloud code team if you would like that sync cloud code. Yeah. Damn. So, so yeah, I mean, this is really good. Obviously I'm like kind of raving about it. And you know, I have other things like sign up for PG and E. So if you can do phone calls from you, that would be great. I do. You don't have done that. Obviously you can do the native. But people have done that with like various other providers. Yeah. And then this is like signing up for the thing where MCP. I really am trying to do like everything data analysis as well. I do think, oh, design to code. Very, very good. So like tears of figma file. I'll take it. And then this is where like a lot of other tasks is like knowledge work. Like replace my manual clicking. But this is no, I would normally use cloud code for this. But because I perceive that you have better Chrome integration, I think you can actually do a better job of this. And I this is one shot in my conference website. That's pretty cool. Like at some point, I would love to like hear how you feel about code in the desktop app. Which is like, I never use which is the the faint team. Same team. So I use the cloud code in terminal, which I perceive to be the default way of cloud coding. So one thing this has, sorry, I'm just like, I'm not here. I'm not here. I'm not here. I'm not sure if people out there want to like hear me advertise my stuff for like an hour. Please do that. Um, this thing is like a built-in browser, which is a thing a lot of products have you send. Yeah, it's a built-in browser. And I think giving cloud eyes and it's like what you're actually working on makes it so much more effective. And that's probably why you're seeing co-work because it can see Chrome. It can like deback the DOM, it could like see things that does make it more powerful. Yeah. So I think my mental model is kind of broken because I only use this code work because I thought you had a browser thing in it. But I understand that the cloud code app or the app version of cloud code does have a built-in browser. I've seen this preview thing. Yeah. I've never used it. In the end, in the end, you sort of get like I hard. Yeah. You basically get the same thing right? Like the, the, the additional skill that you're describing is draw this better if you can see what it's working on. Right. That's, that's so like the summary here. And like whether it's using your Chrome or it's just like making up its own little like browser doesn't really make a bit difference because either way it's going to see what it's working on and that just makes it much better. And then you don't have to run QA for your cloud. Why doesn't it pick up my existing cloud code sessions? Because I, I mean, obviously I've used cloud code but. Excellent question. Don't have a good answer other than like we're honest. Seven. Yeah. This is what the only I'd keep us. Okay. Cool. I don't have other, like I just, I do want to expand people's minds and also maybe show people if they haven't really done it. But like I think it's very interesting how I sometimes use this more than I use, I mean, I use D. Right. Yeah. And I use, I've used like all the other agentic browsers and, and for topic, didn't have to build an agentic browser because you just had cloud core work and, and it's enough. Yeah. I also think like maybe integrating with number of excellent browsers out there is like Kurnion, my personal priority listed a little higher than like trying to rebuild a browser home scratch. Yeah. You know, never seen ever. But I think going back to this idea of like we want to plug this into the equivalent to existing workflow. I think our goal is actually to not replace any applications you have on your computer. But instead like work really well with a new workflow. Like the new one. Yeah. It seems like nowadays especially on the browser, most of the innovation is like user economics is not really like the underlying browser engine. So I feel like the cloud, it doesn't really matter if it's like the, or Chrome or Alice, whatever. Yeah. We want to meet you wherever you are, which is like, like obviously I would say that, but it's also just generally true because I don't want to streak my potentially user base artificially by saying, okay, like I'm going to start building for the people who are willing to switch browsers. Right. That's such like, you know, like many lawsuits have been filed over who gets to have read the browser. And like a lot of money has switched hands over the question of like which browser is default and which search engine is default within the browser. I just want to build for, yeah, I want to build for swix essentially. Like I want to, I want to build for people who have a number of annoying tasks that they feel like maybe clock or do that could do it for them. Yeah. What do you think about skills portability? I think there's been one thing I use another thing called zo, which is kind of like a cloud computer plus agent. And I have a skill to add visitors to the office. Yeah. So whenever somebody has to come in after hours, they need to check in downstairs. But I want to like text the thing. So it doesn't really work in in co work. But now that skill is in the zo harness and it's not in my co work thing. And then if I make a change, I got to think them, how do you see that going? Like I see memory as like cloud personal. I don't necessarily want my memories to be crossing. Yeah. But I do want my skills to be cross agent that I use. I think with empty piece people did the same thing. It's like, oh, I'm seeing gateway, empty be registry. I don't really know if that's like a business. So I'm curious like if you've had any thoughts in the area. I think for me, this is sort of why I go back to the really basic primitives for our skills are file based instead of like this complicated thing that exists in some way that is like super proprietary. I'm really leaning into the idea of like it's all just thousand voters. And that makes it very portable on a son right. We do have skills as part of this container format, which was just called plugins. And plugins are available both for cloud code and cloud code work. The same format. And you can install plugins. This works in code work today. Basically say I'm going to add a whole like just to get up repo as a skills marketplace or like a plugin marketplace. And that's how we're doing part of the ability. I think we have a lot of room left to grow. And how do we make it easy for people to know that they can write skills? How do we make it easy for them to just like share a skill with you? Because obviously all the words are just said, right? Like I'm losing most of the knowledge work I based out there. We're in stark for saying, oh, you can connect to get up repo. It's not exactly how most people will end up working in like a general knowledge work space. But I think there's something there. And another thing that's there that I think has not really been properly explored is the combination of which part of the skill is very portable. And then which part of the skill is like very personal to you. Right. And I think that's something we haven't really saw for years in the industry. And it's like, which is something we want to introduce more structure to the skill or have always have like public skill, private skill pairs. Yeah, kind of. I think that's like the easiest way to do this is what we do, like use string interpolation or something, right? Right. Insert username here, insert like phone number, insert like known folder locations, that kind of stuff. That's probably clunky. That's why we haven't built it. But I do think someone is going to come up with like an interesting way to keep everything well like about skills. The portability is just a file. It's just marked down. It's just text, obviously. Like a text file, which is the complete lack of structure, which means you don't need any kind of. the right skill. Just like explain it to Claude the way he would explain it to me and Claude will probably get it before I would. Right? You just like for booking a flight. Tell Claude how to book a flight the same way we're telling somewhere I just started about me yesterday. But combine that with a very like personal thing, maybe we'll stick with the booking of flight example. I don't actually think AI should be booking flights. I think the tools we have is yes. Yeah. Finally, somebody says it. It's the default demo that everyone was making. I'm like, I need a guess like booking demos. It's not a good showcase. Yeah, I just want to book a flight myself. But I think there's a lot of things that have a personal and a non personal component. And that's me where people reach for flight booking because some things are very universal. Super flight is usually better, right? Like few people try to book the most expensive flight. And then some things are quite personal by like what time to prefer, which see to prefer, which airports to prefer. Combining that and like a school from it that is actually portable, compatible, easy to understand for people. I think that would be very exciting. We're just having to figure it out. Yeah. I think the tech part, I think everybody by now has some sort of like cloud file thing in a drop box, school drive, whatever. So it feels like in a way, it should be like Simlink my skills into all my agent harnesses. Yeah. Just keep those in sync. Like we have internally this like valuable tokens repo, which is like all the commands of agents. It's a good lot. And then I built like a T UI where you can start and be like, you know, install this command and this three sub agents into this agent and this folder and just copypaces. It doesn't do anything literally like CP the file into that. But if there should be something similar, we're like whenever I go into a new thing is like, Hey, here's like the link to exactly the cloud folder and just bring down these skills into this. Like today, it doesn't quite work like that. Like if I install a new agent, I cannot have to like copy base of the skills. And I don't even know where they are. Yeah. It's like the big problem. It's like, where do I find them? Yeah. So I'm curious like in the future, like that almost feels like my personal productivity thing will be my skills. Yeah. It's not really the product that I use. You have everybody has access to the same product. But today there's that just looks like copypacing and the else. I think so many things. I really like thinking about agents and other ones. Just like another coworker. So many attempts have been made to build documentation companies that are like, we're going to solve for you to commendation problems. I myself like spend a little bit of time working in notion, right? I'm like deeply familiar with the concept of let's get everyone on the same page. Right? We basically saying there's the one all your agents to be on the same page about your preferences, about the skills, about the way they ought to work and like how they ought to execute. And I'm not sure what the right thing is going to be. If it's going to be some some company that can say, all right, we're as an independent body. We're not trying to like push into any particular product. It's our job to be like the skill authority and we provide, I don't know, we're going to be the drop box of skills. And we can just similink us into all the products they want to use. I'm not sure that's going to be vital business. But as an idea, it will be cool. Right? Yeah. Yeah. I think so many things are just going away as businesses. It's like, oh, I'm not even asking somebody to make a product about it. Like, yeah, I want a person like now. And there's things like you said, it's like you almost want to skill and then interpolate it between personal and work. So if I'm booking a flight for work is different that I'm looking to fly personally. Yeah. In some ways. Yeah. But like a lot of the scaffolding is the same. You know, I mean, as an engineer, I will tell you like, you know, technical person to technical person, I will just be like similinks. Well, that's what that's what I do. We call that MD and agents that MD is just the same as how a simlinked. And so it's like that works. But it feels like, yeah, I don't know. Maybe we could always go one level up. You can always talk or co-work your problem and then co-work will solve it for you. Just make the similinks. That's like one way to do it. That's true. That's true. All right. Everything is called co-work. Potentially, spacey question for both of you. Which of these industries will go away? Okay. So what Felix was saying before is interesting. There's busy like the short term pressure of like, we need to turn these tokens into valuable things, which is I should build the last mile product that harness the model. And then there's the question of like, long term, which ones are gonna still be valuable. And I think you're kind of seeing this today with like, you know, the coding space in a way. It's kind of like, everybody's moving up and up in stack because you need more than just turning tokens into code. I think search, like enterprise searches, kind of seeing the same thing like with a Glean and like all these different companies is like, I think of the day of co-work as the one doing all the work. The search itself is like such a small part that like, I don't know if I'm really gonna pay that much money just to do search. It's almost like everything is like a co-work vertical. So like, how much can co-work first-party support? And how much can it not? I think for a lot of these things, the planning thing that you were showing. The vision? The planning. The planning. Like, that's one thing where like most of the value, the decisions provide us like, they're better planning for specific tasks and a better tools for it. Yeah. But I think the models are now moving in that direction and they have the right harnesses and they're on your computer. So for me, it's almost like, if the end customer trusts your startup to be the provider of that task result, then I think that works. This is something that, this is a strike that we're working on. Yeah. I think, look, I'll tell you this. I don't think I'm the best person to actually estimate which industry is going to be hit the hardest. But I do think that at Anthrophic, as a group of people, we're deeply worried about the impact that the tools are going to have on the labor market, especially for like junior employees. Because I think it's only honest to say that when we talk about automating a lot of the work that we personally find annoying, that we maybe think some of the best use of our time in a lot of industries that kind of work would have been given to a junior entry level. And I think it's only right to be really worried about that and like, worry what that's going to do in particular to people like enter the shop market. I have a solution for that, which you make them, you create, simulator jobs for them. Okay. So this is like half joke, half true. So if you think about software engineer, when you're like a junior engineer, you work like one, two, three years. And in those three years, there's like maybe like an unfold moment where like you really learn something. And then a bunch of other days were like, you're not really progressing. Yeah. I think now we can use AI in these models to actually like shortcut these careers and almost like simulate the early years of your work and like, you know, make them like super dense in like these learnings. As I, hey, we're working on this feature, which is like a distributed system and you need to learn this thing. That may take three months out of company. And so you take three months. Here's like we're just simulating the whole thing. It's actually not a real thing. And in one week, we kind of speed run through the whole thing and you kind of learn your lesson from there. And we kind of repeat that. And like one year, you basically get like three years worth of like projects and experience. Yeah. I think it's harder for like things like sales or for things like, you know, marketing because you don't really have a way to get the feedback group. But I think a lot of it, it sounds kind of silly. It's like you're making the new effect job. But it's almost like you go to college, right? People pay to learn how to do it. And this might feel similar where it's like, hey, we have the Jane Street simulator is like, you want to come work at Jane Street? We'll just put you in the simulator for like three months. And you'll come out of it. It's like, you know, I'm ready. So there is an aspect here. I'm not an expert enough to like actually know what, what is going to happen to marketing or legal finance, right? Like I don't work in those jobs. And I don't think I should talk about them. But I am an engineer and I think I have a pretty good idea about engineering is like. And I think one thing we're sort of seeing is that there's a company and also it's the public, we're like, deep, we're at entry level. But we're also seeing more senior engineers accelerate it. If you like, they're more productive. They actually increase the value that provide. And the thing that I'm thinking what a lot is the fact that even before all of this happened, I've always had a lot of respect for the University of Waterloo and the new grads that have joined my teams as from coming from the University of Waterloo always felt like more ready than new grads will literally spend their entire time at the university regardless of how good, but never actually had a work inside an environment. We have to ship things that eventually will be used by users. And I'm German, I like initially meant to German University. I think the the like information systems programs there tend to be very theoretical. Right? Like often give people the example of like trying to become a doctor, but you first have to do four years of biology. And as a result, when you get a new grad, you sort of have to teach them what it's like to actually build products and to work in a company and like work with other people. And like some people will have a different opinion and like, how do you do all of those things? And the University of Waterloo, it seems like they just spend half of that time, I don't know if it's true, but I think it's a year, right? They spend so much time part of your job, a curriculum to do spend a year in internships. Yeah, they just like go from company to company. They show up on your team as like a junior engineer who has been to like 20 companies, not really, but like it seems like a lot of my new grads have also briefly worked at Apple, Google, Tesla. Yes. And there's a common meme where they like collect all these logos like Infinity Stones. But and they always put you on LinkedIn and it's very unclear that they're in intern. Like yeah, yeah, exactly. But it does actually make them so much better compared to other new grads. And I wonder if that's a useful model maybe for the future when we all say after like crunch down the amount of time you have as a junior employee because the value you have as a junior employee is going to like be impacted. My sort of pro young people take is that you're more higher neuro-classicity, you can learn more, give less pre-existing biases. And what I assume is true for you, what OpenAI often says is that actually is the younger, like, first grad engineers that use codecs or their coding stuff more innovatively than the experience engineers who have a set and preferred way of doing things. Yeah, as I talked to people, I had some of my experiences. Yeah, so maybe you're more AI native and therefore you get cut. But I mean, I think the problem is you don't need that many of them. I mean, and for topic is on the record of saying, we do believe that the impact on the market is going to be sizable and we do not think that people overall are ready, right? And we do actually think we should probably talk about it as a society much more. Yeah, I'm not sure that I'm like the individual that can add anything you use to play. But I think as societies with economists and governments that need to wrestle those questions in a way that is probably more meaningful than me wrestling with them, we're probably not doing that enough. Yeah, we'll try to educate. And then I think also just releasing frequently as you guys do, or probably maybe too frequently, is helping people to adjust over time, rather than one big bang thing. There's this gradual takeoff that people are living through. Yeah, yeah, but I think a lot of us wondering what kind of we actually have full takeoff. Like what point is there more sort of expecting this big bang moment where things will accelerate so quickly that it becomes a self reinforcing loop. And at that point, it's sort of like off to the races and there will be no more slowly catching up. You're not just half clad being so good at everything. Yeah, it's when co-work is training models. It's when it's looking intensive forward and exactly waiting by his eyes and training things. We can all debate how many years it's the way. Some people make it better, maybe it's 10 years away, maybe it's a year away. I'm not entirely sure where we're at common this time, but I'm not entirely sure that ultimately it matters all that much, whether or not it happens in four or five years. If we have a decent one, certainly that's going to happen, it's probably something we should wrestle with. I wanted to talk, so by the way, the scheduled task complete, there's the clean my desktop task complete and it did. It organized by file type, which okay, but I was trying to get it to do more cinematic, read the file, understand what it's about, grouped by the topic rather than the file type. I mean, you can just follow up and have it do that. So here, it is proposing that, right? Yeah, so it's got some topical things, but I could probably do better. I probably need to give it a skill to read video files so that understands the here's how I like to. Honestly, though, I see they're using Open 4.6, right? My recommendation for people is increasingly don't worry about it anymore, just like tell it what do you want it to do. And it's probably going to figure out a way to do it. Okay, it might not be the way that you like, necessarily the way that you've gone about it. Videos deeper. But we're outsourcing, organizing. Oh, well, it is, so let's fight. Yeah, yeah. I'm honestly like so curious what cloud is going to come up with. I'll kick that off. I wanted to also just talk about the overall, you know, you talk about data analysis, you talk about like your personal finances. You also said, which by the way for us is very timely tax season, right? Like, yeah, use cloud core for tax season. It is not responsible for any mistakes, but might as well, right? Like, it's free knowledge work for you. Yeah, so I just like, I think cloud for finance is a big deal. And this is definitely like in that mix. I wonder, is it like, do you, is it a separate team? Do you talk to them? How important is it? Right? Like, you can also natively output Excel files now. Yeah, just talk about the finance effort. Yeah, I'll care about the verticals quite a bit. So we do have a dedicated vertical steam, also a dedicated enterprise team. And those business engineering, not sales. It's engineering. Yeah. Yeah. It's engineering. So we do have people who sort of come to work every single day and they ask themselves, how do we make co-work extremely effective for people in those specific industries? How do we make it easier for them to understand? How do we make it easier for them to pluck into this and like sort of get the same value out of it that software engineers get? I think it's not real surprise that software engineers ended up being sort of at the forefront of the entire AI moment because so much of it is this like group goldberg machine desk where like we're already used to automating things, right? Like it's part of our job. Yeah. So we care about it quite a bit. I think it also like really matches what we see cloud being very good in as a model. I think it provides tremendous amount of value to those customers in particular because we can do so much with the amount of data they have. Those are like data heavy industries. Their industries were correct and as matters quite a bit. So for us, if I've used it to analyze my business, I just can't show it. So it's too sensitive. I had a similar question about what taxes. I did read about the fact. I did read about, oh, co-work is doing my taxes. This is honestly incredible. And it's like annoying. It's like, this is so cool, but I'm not going to Twitter is maybe not the audience. Then you still like see my text returned. Yeah, but here it is. It's reading on the videos. So it's like, yeah, it's getting more. Yeah. How did it actually do it? I actually curious. Oh, usually it just like takes the screenshot and then it reads the screenshot of it by vision. So this is what I do for my zoom upload thing, right? Because I have paper club sessions that I need to upload to zoom. And I wanted to automatically title them and do show notes and everything. So just take screenshots and try to try it's best. Yeah, it wouldn't probably benefit from transcribing, which it's doing by it's operating my pure vision now. But it's good enough. Yeah. And then I do have to call out to nano banana to do images. So unless you guys do images for me, I have to call the people. We were images. We're aware. We're aware. It's just so fun for me because this is the thing that I'm increasingly doing. Like, increasingly curious about cloud creativity and like for coming out, what is great cloud approach to like some problem? Yeah, vision for everything is like the superpower, right? Like, you know, and computer use, you guys were the first to do computer use, right? And when it was launched, I was very unimpressed. I was like, it's slow, it's unreliable. It's the why I have so much better. It's because it's one year ago. Yeah. I know. Like, it was barely usable. Yeah, I remember I was very usable. But it's a wild, how much better things have gone yeah, yeah, over that one year. We went to the orthographic office because of the launch event for computer use, like there was like this hackathon. Yeah, like nobody hacked on computer use. But I did see, I don't know if you were talking with me saying that, but I did see briefly that you do have like an automate mech OS, MCB server installed, right? You said ever. Sorry, which one? Where? If you go to your settings. Oh, settings. Okay. We're sorry. This one. Yeah. Yeah. I noticed that in your connectors. I probably said it at one time, but I don't use it actively. Okay. The Maxwell is automated. Yeah. Yeah. So, so yeah, this one, I really wanted to just automate everything in my thing. I didn't find it super reliable. Okay. Why? No, no, no. Claude is much better writing Apple script and executing its own Apple script than relying on these third party tools. Yeah. So I've increased, I initially installed IMCP and like all these other FCPs that people built. But now I don't use any of them anymore. Yeah. Just like Claude, right? It's own thing. Yeah. It's going to be more custom made. We keep going up the stack. But at least in computer use is like a fairly interesting area to me. And it's like also interesting in the sense that I think we're far away from other thing with far away from clapping very effective at like using your computer and not just a theoretical computer. What's the relationship between the user and the computer? Like that? There were some tweets about how huge some of the VMs that Claude co-work creates are like 12, 15 gigabits and people playing. But at some point is like if you're using the computer, you're taking an action. It's just your computer and I'm just looking at it. You know, it's like I think that's why people like the idea of like the Mac Mini and the OpenClaw or whatever on it because it's like it got its own home, you know, it's doing its thing. I'm doing my thing. I think there's some kind of like not like race condition, but it's like, okay, if I kick start the stask, now I can't really use the computer. Yeah. You know, because Claude co-work is doing things on it. And it's kind of awkward. Like yeah, I'm not sure. I do think it's a super interesting area because I can maybe tell you like some of the things I thought about that I think are actually a bad idea. So when when we initially started working on co-work, I did have some dreams about what would it look like for Claude of its own cursor? It could be cool, right? Like it's a computer. We can write code, we can touch everything. Like who says that computers need to have one cursor? We could do a second cursor. But then actually breaks down quite a bit even if you go and like present cool dreams to both Apple and Microsoft, you're like wouldn't it be cool if it breaks down quite a bit because so many of our models on the computer are built around this idea of like there's only one thing you're working on it. Yeah. There's like a foreground app, a background app. Claude and Chrome can work in the background, but it's like within one application, but it's the operating system layer that is a lot harder to implement. So I'm still grappling with what does it mean for Claude actually act on your computer? It's the right format for Claude to have its own computer that you set up and maybe every other day and you're like zoom in and you play with it. Or is the right format for Claude to just like wait until you're stepping away for a little bit and take over while you're gone? Or is the right move for Claude just like a horizontal computer in the cloud and like whatever you want Claude to do, you have to set up yourself. There's like a number of different options. This is a thing I think about a lot like what is the relationship between you and your computer and you and your data on the computer? Because how intimate that relationship is kind of depends on the tool and the thing that you're looking at, but quite comfortable sharing some things, very uncomfortable sharing other things. And I think whatever product is going to be successful will have to deal with those different things, but you probably even if Claude was capable of making a determination, would you want Claude to make that determination in the first place? It's tricky, but I agree, because I'm going to be a little bit more patient. is like, it's more than just privacy. It's like almost intimacy. And it's like tricky to reason the bar in a way that will make everyone comfortable. - Yeah, I could see a virtual box, like actual virtual box app, where you run the VM and then you have like a screen within the screen, you know, you can put in the background, but then you can like jump in the screen. And like-- - You're in that not a bad idea, yeah. - You know, like, I mean, I used it, you know, people used to do it, virtualizing like Cali Linux, you know, Windows machine. - Yeah. - And like, you just jump in and then you would jump out, but it's like, it's not like a dual boot, it's like within the thing. The problem is that you need twice the amount of RAM, twice the amount of, you know, it's like, it's kind of taxing on the machine. But I think that would be cool. Kind of like see, you know, the little card window, I can see it's desktop, like, how cute it is, clicking around things. - I was gonna bring out, he's the original machine and the machine guy because he has the Windows, Windows 95 project, where's the Windows 85 project type? - There's probably someone on my guitar. - No, no, no, no, no. It's like, it's the first thing you see is this one. - Nice, yeah. (laughs) - Yeah, exactly. - There was honestly a very fun project though. Like, obviously I didn't, I should say this, just so that no one gets the wrong impression. I did not write the actual, the actual, obviously I didn't build Windows only five because I was a child, but also, I did not build the actual engine that is capable of like, simulating a Lexi-36 processor and JavaScript and Ratham. That's a tool called V86, which is very cool when everyone should try. But this came out of like a debate we had at work where people were like, they often are in the interdibating the merits of electron and whether or not we should be building software in JavaScript, yes or no. And I still am very upset that I can run all of Windows 95 in JavaScript and launch Microsoft Excel inside the virtualized JavaScript Windows 95 machine and do things that I can do that entire chain faster than they can do a lot of other things than like traditional SaaS applications. This is sort of like a performance rampage that I went on. So I'm mostly built as a joke for some of my colleagues at Slack, this took like one night. What? But then, it was not hard to do. It was all the hard workers in V86. Like it's got to the repo, it's kind of say like 90% of this work is done by a guy who goes after the, by the name copy, his name is Fabian. Yeah. Cool. I think you're kind of back on the Windows Brian because you're building on the Windows support. I thought there were some really cool technical stories to tell and it gives people an appreciation of like, well, here's how hard it is. And here's how important, how you invested the sandbox. So maybe this is like a good opportunity to talk about some of the details. Oh yeah, the VM, honestly, it's like so cool. There's a lot of things we dislike about the VM, right? There's a lot of things that are real trade-offs and you want to know why I'm in those trade-offs. And you're right, there are a lot of people write me like, hey, how come Claude is taking out 10 gigabytes? I could say on the point of thought, actually taking out 10 gigabytes, it's just like a way that mechos displays bytes is like wrong. But the way we actually write it to disk is by we collapse the empty space and the image. So it's not actually taking out 10 gigs, but that's a technical differentiation that's for not going to have a trial. To me, the, the, the outcome is it takes too long to start. Yeah, it's like 30 seconds or time, sorry. I don't know. Oh, it should be faster than that. Whatever. Here's what we'll be telling about, it feels like 30. Yeah, like even either way, like whatever it is, it's going to be, it's going to be slower than just running Glow-Cotor-Wack-Linear-Gopritor, right? So the trade-offs are real. But what we're doing on Windows, we're using the Windows Windows host compute system. It's the same thing that WSL2 runs on, like the Windows sub-system philinics that I think a lot of developers appreciate quite a bit. Yeah. And it's, it's pretty cool because we sort of like have to separate it out, which system space, the switch on machine runs in, and who gets to talk to that virtual machine? Because obviously you give this virtual machine a decent amount of power. How do we optimize, not just the connection between the two systems, but also how do we make sure that random other application doesn't get to talk to Cloud inside the VM? Would you send pretty interesting things? Last week we started writing a new networking service and networking driver that optimizes how Cloud talks to the internet. If your company is doing like weird internet things, like the patent inspection, like the, like, you know, taking your part as a cell inside your company, I think there was probably like a very small, easy version to build off-co-work that is much simpler, but also breaks on most users' computers. And this one is quite nice because it works on most users' computers. And the default example I was going for is, I really want this to be highly effective on like a, on like a machine that most people pick up. And that machine will probably not have Python, it will not have Node.js. And even if I just take away those two things, Cloud is going to be so much less effective from your computer. So what do you do? You don't even, I mean, maybe require people to install Node and Python, like you know, in front of like a, what is the future like, like without a VM? No, no, no, so like you say, right, let's say our get machine is whatever it's a default spec windows that talk. We do this. We're just quite cool. So on Mac OS, we use the Apple virtualization framework, which is pretty solidly optimized. Like it's good stuff. And it's a simple API call, right? It's like super simple. I saw the code recently. I don't like that's it. What the fuck? Would you once you start like shipping production code on it, you start adding like all of these edge cases, you're new. It ends up being a little longer. But I think Apple really cooked with a virtualization framework. And it's very, very good. It is very fast, it's very reliable. And same on Windows, the host compute system, I think WSL2, as well as maybe one of the diamonds with them. Who knows it's like one of the few things that developers universally rate about is very, very cool. And like hooking into the same subsystem makes it a lot easier for us to say, we don't really care how locked down your computer is. Maybe it's like your employer's computer and your employer has decided that you get to install nothing, not trusted. But it's true in a lot of environments, right? Like even at Anthropic, our IT department controls work on a software install. It's like a pretty common experience for many companies. And this gives IT departments a decent amount of-- it makes their job so much easier. Because we can say you can separate out Clots computer from the users computer. And then for Clots computer, where you probably care about its data loss, you care about a potentially hostile actor, you care about maybe data being exfiltrated. And once you control the network and the file system layer, you don't really care necessarily anymore. The cloud might be writing super easy to call Python scripts. What worries you about the fact is that once you install Python, now anyone can do anything on a computer. But once you put that in a VM, that risk really goes down. Yeah. So that's why we jump to all of these hoops. Yeah. I think you had a different tweet about this. But it's almost like people have also approved exhaustion. It's like you can't approve every single command. Sometimes by default, some of the CLIs-- I think even early cloud code, we have to approve every single command. Yeah. And like it's-- there's this sort of dichotomy between either approve every step or dangerous these permissions. Yeah. And actually sandboxing is kind of like the middle ground. Yeah. I do think it's maybe on us as the industry to come out of something better than, oh, this is super safe as long as it doesn't do anything. If you want this to be useful, then you have to approve every single step of the way. And like, computer uses a good example. The only way to make computer use on your host, super safe, really super safe is probably if you approve every single action. Like, models like I would like to type the word L. And you're like, OK, that seems fine. I know which like cursor is for-- yeah, it's not automation if you don't delegate. Yeah, exactly. You need to probably delegate. You need to be able to like delegate and walk away and trust that this thing is not going to like mess automatically. And I don't even think we need to build perfect systems. I don't think we need to wait for like 100% model alignment. We can rely on the same Swiss cheese model we've used in the industry for a long time. But I do think we need to like universally maybe eventually invest more. And therefore we do. We need to invest more in systems. But we can say you do not need to approve everything. Speaking of Swiss cheese model, I just wrote the thing about this. Oh, cool. Yeah. Yeah. Yeah, super cool. I mean, yeah, it's weird how like, I guess, usually, I think safety and security is kind of like a boring word to engineers. They're like, just give me unsafe to me. Unsecure. But I think achieving the right thing, like you're going after a consumer slash prosumer. Yeah, yeah, talking about those. Kind of like both. I think I also want to capture people who would have no trouble using Cloud Code like yourself. But still find them maybe just convenient, easier. You're like, oh, cool. That's like the drill is on the right. I can add to it. Those things are just easier to do if you have to. But this is like clearly the knowledge work side. Yeah. Cloud Code will clearly capture the development workflow. But like, I do think like you have to sweat this safety and security details in order for people to trust it. And like even Cloud and Chrome, like having the whatever API uses to do the background thing, that's the only reason I use it. It's because otherwise I would have to just get a separate machine and just run it. Run it. That sounds super annoying. Yeah. I mean, I currently doing it. And I think also as developers, maybe where we are more risk tolerant, but we're also just like accepting, we are more risk tolerant. But I think we also just have like, I don't want to say arrogance, but like sort of the trusted if like the really bad thing happens, we can probably fix it. I just tell Cloud to like, I check with me before doing any irreversible action, like sending an email or doing it permanently. It's good enough. But like not even Cloud, I mean, like simple things, such as NPM install. And like we're all running NPM install with full user permissions. And if it wants to like read dot SSH, it will crazy that that is the default kind of. Yeah. I know I agree. I agree at the fine. Like I'm obviously doing it every single day. No, right? Like I think obviously NPM and get up to have like done a pretty good job, maybe over the last couple of months, like clean house and come up with more specific tokens. But generally speaking, I think it's engineers. We've always been a little bit more risk-clarer in. And if you do a little bit of introspection and you ask yourself, is that how we should be doing things? You might not always come up with the right answer. And I think from models to like my approach, like I'm not going to, the safest thing is to do nothing. We do want products that are quite capable. But to the extent possible, I don't want to ask you, are you okay with the script? Because I kind of believe that once it starts becoming a part of your workflow, you're probably not either, either you don't have the skill to understand whether or not the spice of the script is safe, or you're not going to read it anyway. Cool. I guess a couple of parting questions. What's the future of clockwork? I think we're still such early days. We're going to keep shipping things that, we're going to keep shipping things that, we're going to keep it a rating of the same, like pretty quickly. But which I mean, you can sort of continue to expect that every single week, there's going to be like a small new feature, if not a big new feature. I'm going to continue probably to double down on your computer, and like making you effective in your computer, making cloud effective in your computer. We're starting to grapple, as we talked about today, grapple more with the question of like, what does it mean? What does your computer mean? Does it have to be the one in front of you? Or like a VM on your computer, or like a computer somewhere else? And then the third thing that I'm quite excited about is, we're continuing to go up this hill climbing on slowly taking users who are used to asking questions and getting an answer, to slowly teaching them to like step more and more away, and that clothe over bigger and bigger tasks and work, both in time as well as in like scope. And I think you can probably see most of our investments on our feature releases, like working both of those things. Like the ability to do more on your computer, and then the ability to do more independently and for longer. There's remote control work for clockcork yet. No, right. Excellent question. Coming soon. I mean, that's an obvious thing if you want to keep betting on your computer. But to me, like, we talk about like people are not ready this year. Like there's no wall. It's exciting. To me, like, what will we be doing differently at the end of this year that, you know, we're maybe not even thinking about this at the start of this year, right? Like, I'm just trying to look ahead as to like what's like a good use case that you're with it was sort of aimed towards. So for example, for the machine learning scientists, it's always, okay, well, I want AI scientists, they can automate machine learning. But like for knowledge work, I mean, I can already, you know, get it to sign up for Google Cloud to mean as a GI. Because Google Clouds. But like, what is what's beyond that? I don't know. I think it's basically the idea that like you still have to tell her to build your script, right? You were still kind of involved. Yes. And maybe a way that felt kind of magical to you, but like maybe to me on the other side as the person building this product still feels kind of heavy-handed. I see so much process that I'm like, oh, let me take that away from you. Okay. Like, how do I just go? I will continue to go, we'll continue to go like further and further up the stack and make your life easier and easier. Oh, here's one, right? Yeah. Watch. You know, I don't care about my own privacy, whatever, or trust, I trust the topic. So just watch everything I do on a normal day-to-day basis at the end of the day. Tell me what you is called co-workable. Yeah, I don't know. I think the funny thing about a lot of these products is that like for good reason, I don't enjoy, I don't feel like my entire career. I've never liked teased too much what I'm working on because I think you should just like, yeah, to lose it. Yeah, build the days and release it and then talk about it. Like I'm not a big fan of that like vague posting my own work ahead of time. Yeah. But the thing that is like always so fascinating to me is like both of you all multiple times today. You've like mentioned things and I'm like, yeah, that is obviously like very obvious. Okay. That someone should be working on those things. And I think we're still in the space where if you look at co-work, the things that we will be releasing will probably not be a big surprise to either of you. You're going to be like, yeah, obviously that's valuable. Obviously they were working on those things. Yeah. And obviously that's going useful. And the more I hit those points, the more our features fit into that category, I think the better it is for us because I'm we don't end up building things that are due hyper-specialized to two difficult on the style. Yeah, I think the hyper-specialized thing is very important. It keeps you like general purpose. It means you're not thinking too small maybe. I don't know what the word is. Yeah, exactly. It's like the whole concept that like a no point if we release, you know, there's no cloud code for no jazz applications. They use React and Ten Stack and only those two things are like, if it's anything else, I know several startups like that. I think that's pretty like I'm not a VC. I'm not an investor. It's like hard for me to predict where the markets go. But in terms of the building box that I'm interested in, the electron is probably by far the most popular thing I ever built. And electron itself is like very abstractive and generalizable, right? Like so many apps running it that I think it would have been hard for me to predict how many apps actually end up using electron. And what would have been even less useful for me to predict this in what those apps do? I just see the rim of bloom coming out of me. That is cool. Like you're a camera in a little circle in the corner. That is pretty smart. That's an electron app. Yeah. Or at least, I'm not sure if it's still there, but it was for a while. Like one password has so many interesting things, right? It's a level of the stack that I'm quite comfortable with. And whenever I give other engineers advice, it's actually that layer that I think is most valuable to invest in because the tools of the layer are not that good. But that's where you get the most leverage for the future in general. Just quick tangent on electron because I always wonder this, have you looked at Tori? I have. Yeah. What's your take? You know, my view is like most things should be Tori by default unless you really need the full power of electron. But yeah, I can give my big take. Why do we ship an entire version of Chromium inside the thing? Like what do we do that? And people ask me this question a lot because it's like very counterintuitive. Wouldn't it be much easier to use the web views that are on the operating system? Wouldn't it be much easier not to have to do that? And the answer is yes. And like obviously I did that once upon a time I did that. There was a version of the Slack app that used just the operating system that views. Did you start the Slack app? I would. Well, team effort and yeah, but I was there. I built the Slack app. Yeah, it's crazy. I mean, obviously get the electron guy to do it. But well, but this is an interesting find like by the time by the time I joined Slack, they already had an app that was built with something at the time called Mac app. It was a little bit like the same app got thing for mobile. It just used the operating systems, web views. And that didn't work for like so many reasons. And they were like, all right, maybe we need like bigger guns. We need to like take more control of the rendering stack. And there's a few things I was mentioned here. I think if you're building a small app just going with the operating systems, that view is perfectly fine. If you're building an app, maybe that doesn't have too many users who will like cry bloody murder if it doesn't work that is fine. The reason to go with your own embedded rendering engine is because, and this is still true in 2020, 60 operating system render renders are not that good. They're just not that good. Both Microsoft and Apple are trying to move away from that. They so far really haven't the only way to upgrade those is to upgrade your operating system. So if you're a sales lack and you have critical rendering bug in WKVBU and some of the other web view options, your only recourse is to tell your customer, oh, sorry, you're too poor, you didn't buy the latest MacBook unacceptable unacceptable to user unacceptable to user developer. So you sort of need to like go down the stack and like find the best rendering engine and put it in your app. Why Chromium even knows very big Chromium is by far the best thing like I often like to remind people the Unreal engine you want to render some text they use Chromium like Chromium is part of the Unreal Engine for same purposes Chromium is very, very good. I think it's like one of the marvels of engineering. It's very hard from where it's ever since go right out of the recording. Most of the people in the city are web developers. It's hard for me to like overstate how magical it is. They can run seat like rendering a YouTube video dynamically negotiating a bit rate, figuring out what to do about your extremely broken hardware driver. Actually, this is a fun thing. You can enter Chrome, call on WackWack GPU. Okay. And if you scroll down a little bit, these are all the enabled workarounds because something is going wrong on your computer. If you're doing this on a Windows computer with like a GPU that is not the most popular GPU, it will be much longer. And all of these are usually just there to make sure that if I say as a developer, I want a red pixel to appear here that that actually happens. Chrome is such a marvel because it works on all the machines that use on my throat and it's going to work fairly reliably. And if it doesn't, they will probably fix it within 24 hours. As he says. So this is the super operating system right there. It works everywhere. Yeah. Yeah. Yeah. Yeah. So a lot of the magic of Electron is honestly just that it makes it very easy for you to show Chromium in a way that serves you exactly in your use cases of Electron. Exactly. Our next interview is with Margin Jason, yeah, who had the phrase like best up OS's are just poorly deep poor implications of the actual OS, which is Chrome, which like actually works everywhere. And this is this is the platform where you shoot apps. I think the wild thing is that I guess engineers who often sort of assume that the platform like the layer below us is like super stable. And they talk to those people in the like out. We're also just like guessing. And I have like a distinct moment at Slack where one of our customers at Sack was in video. And for a while, I really put GPU developers on this pedestal in my head. And I do think they're still probably much smarter than I am. But I was like, hardware engineers who built the chips, who then built the drivers. Their work must be so much harder than mine. They must be very good. And we had like one bug and Slack, well, like if you had a YouTube video in Slack, it wouldn't quite render why, like, what have this weird artifacts? And that ended up being a crime. and I ended up on this giant thread. So I got to see a lot of the source code and they also are just like, common to do, we don't know why this is weird, but if you flip this bit, things work. You know, this is just like happening with Everlay or the stack. Maybe the end of year of A.G. App Rediction is that cloud can build Chromium. (laughing) You see, you see, you laugh now, but yeah, you know, someday, it's, it's drowning. You get pretty good, like it used to be completely useless. Mostly just like overwhelmed both with the whole hyper specialized tools are inside the Chromium repo. Like for a long time, the Chromium is sort of reinvented or the tools because none of them are capable of ending Chrome. I think the EGI moment, I'm kind of waiting for is, at what point are we gonna say, Electron is probably not gonna be necessary because you can just build fully native apps. The Swiftie, yeah, like not just Swift, because this is one thing, like it's pretty easy, if you, I think our current models are quite capable of taking an Electron app and replicating its Swift. Are they gonna be capable of like building an app that is actually more performant, is less memory, all of that stuff. And it's gonna go into the same hyper optimization that developers have done for a long time. We're not quite there yet, work in like, client, even our best models at a thing and say just replicate this in native code, make no mistakes, ultra think, right? We're not quite there yet. Not just think it's better. Today, everything is back. Yes. Okay. We'll get on edge, we'll think for like days. Just a pretty long time for work. But he worked on what you think for days? Yeah. Why? It's just a front. I'll let it all go into it. Yeah. Another question I had is like, co-works. So if I have my cloud co-work, like what's kind of like the multiplayer mode, I think Sub-Igans is like single players split up the context. Yeah. And the multiplayer co-work is like, my colleague is some file on their machine that I want to know about, or I want to know how their task is going to then update my thing. Like, is that interesting? Is that something that makes sense for you to build or for like? It's like super interesting to me. It almost goes back to like some of the scaffolding room. Like, okay, are we going to be end up, are we, will we end up building scaffolding that will just go away? And like a question I have here is, at what point do we just assign these things, like their own Gmail account? We'll just give them their like Slack handle. And then it will just like use the same tools we humans use to interact with each other. You mentioned our refinance people. They've been working pretty hard on very good office integrations. And I think for a while, we built so much tech around, cloud-leaving, useful comments inside Google Doc. And now it just does it, just like leaves a comment in your Google Doc and that's how you interact with it. Maybe like the similar thing where I still have open questions around what is the best interaction mode? Is it for us to go something super custom for co-work agents to talk to each other? Or is it, okay, let's just jump straight to the finish line and say, well, we're just gonna give this thing, if you use Slack at work, we're just gonna give this thing a Slack handle. And that's going to be the way it's like multiplayer capable. - They communicate with each other. - Yeah. - Like, you know, as a fun project, I build this thing called PiQ, which basically takes any repo and the Pi agent, coding agent, it puts it in a VPS and then there's a public webhook where anybody gets a meta coding task. - Oh, then there's a dashboard in which you review the task and then it's like, Pi, Pi, Q. (upbeat music) - Yeah, you basically get all these tasks, anybody can submit a task. And to me, it's almost like in the organization of the future, it's like the salespeople are talking to the engineering team that is talking to the marketing team, to the product team. And all these co-work are going to like, queue up decisions for other people to approve in a way. - Yeah. - And I'm kind of curious what that looks like. And like, how do I give my co-work the ability to both approve tasks without asking me? - Yeah. - And how do this I, which one I need to review? - Yeah. - You know, because for some of these things, it's like, you know, you want to change the color or something that's kind of like a branding decision. Or another one is like, hey, your thing is just broken. It's like, this is like how you fix it. Cloud can actually review, but or not, that prompt matches what it's trying to do. Today, everything is still very, it's like multiplayer with them, the single player, you know? I get up and up many of them. But like, how do I get multiple people to hand off to each other things using their particular context? - Yeah, and for both of your co-work, so like talk to each other, right? - Right. - So yeah, we gotta have us, so today, can you like, have you, you know, or? - Yeah, this is like, I know we're like running out of time here, but like, we previously talked about sharing skills and I did have this question of like, what if your co-work would just like, ask the other co-works if they have a skill for this task? Doesn't know those things we could do, all right? Like, okay, so skill transfer. - Yeah, like, and again, this maybe, the Dp. (laughing) This maybe goes back into the territory of like, building something very powerful and building something creepy, often goes hand in hand. Because I could tell from the reaction that my fellow engineer said that this is probably not what we're gonna do, but like, we have Bluetooth LE, right? Like, this computer can figure out that it's sitting right next to this computer. So you're probably working on the same thing. But you see that in co-work probably not, but there's like, I think really creative solutions to problems that we really haven't tried yet. - Yeah, yeah, yeah. - Excellent. I guess the last thing is the end-thropic labs. I always have this mental model of a model lab versus agent lab and this is basically end-thropic's internal agent lab, which Cloud Code is now under, right? It's part of the whole org. I mean, people are still fungible, right? Like, okay, this is just, yeah. I don't know how real this is. - I don't know. - No, it's a real team, it's a very important team. The last team is primarily working though on things that you don't seem public yet. They're trying like really wild out their ideas that seem quite improbable. The mad science, but you're officially under this thing or now we're, we're, so Cloud Code is, but now Cloud Code is like a fairly big group where I actually know many people got like, like I remember yesterday coming into our weekly COVID meeting and I was like, woo. (laughing) It's a lot of people here. But we still have a lab team. We actually made the lab team a lot bigger. Mike just joined the lab team as an IC, which I think is very cool and very fun. But they're working on things that you have not seen yet that are extremely out there and probably half broken, right? Like the sort of the idea of a lab team is that it should only work on things that make really no sense for anyone else to work on. - Okay, well, looking for exciting things from there. But thank you so much. I know we're out of time, but I appreciate you're joining us. I appreciate Cloud Code, everyone go use it. It is the closest I've felt to each other this year. - That's so nice of you to say. - Yeah, you're very much. - Yeah, thank you for your time. - Yeah. (upbeat music)