How Anthropic ships the #1 LLM in the World — with Katelyn Lesse and Angela Jiang

[00:00:00] Hi. Today, we're starting something new. We're launching coverage of Anthropic for the IPO. They just filed their S-1 as of today, and so we're releasing a series of podcasts leading up to the IPO, which is expected to be in 2 weeks for $2 trillion, and it's one of the fastest-growing companies of all time with a reported lifetime cash raise of hundreds of over $100 billion and a lifetime burn of less than $10 billion. So, um there are a lot of efficiency stories to dive into there, and I think that's something that we want to cover. How Anthropic is simultaneously shipping the strongest models in the world while also doing it very, very efficiently and building an incredible business. So, I've pulled out, you know, a lot of you know, if you've been listening to the podcast, that I have a former life as a financial analyst. And I pulled out some statements from the S-1, which I thought would be interesting to give context to today's podcast. And we're just trying this as a new thing because I feel like some of you [00:01:01] want that personal relationship with us talking directly to camera rather than talking to our guests, which we've done for 4 years. So, the first thing I want to highlight is that Anthropic is an extraordinarily efficient company by human counts. And that's maybe obvious because it's an AI company, so obviously they're using AI very well. Um I think the degree to which they have changed as a company over the past 1 year is not well understood. So, uh trailing 12 months revenue per employee is now 6.6 million as compared to 1 year ago, it was 1.4 million. So, they've grown almost five times the revenue per employee. And I think that's something that we need to dive into more because I think any business owner, any engineer, you want to increase your revenue per employee, not just to increase profitability, but also to increase the amount you can pay your people. Um this is obviously tracked by single thing increase in revenue per dollar of [00:02:01] compute. It is notable that basically in Q2 of 2025, Anthropic was basically losing half a dollar for every dollar of revenue. And now is making $2 for every $1 in rev in cost. And I think that's an interesting uh, analog analogy of the like how enormously huge this inflection in Anthropic's revenues and business outlook has been over the past 1 year. Like this year was extremely critical for Anthropic to execute well and it has done that. And I think we're going to basically going to drive deeper into this and tell that story as much as we can from the podcast external point of view. Um, I the one consequence of this entire increase is that Anthropic has just reported its first profitable quarter. Like not adjusted profit, which is profit after we take out all the things that cost us money. No, this is generally accepted accounting principles GAAP profit uh, of 19% and that is enormous again, completely flipping from [00:03:01] 1 year ago. Um, more than flipping. Like the 1 year ago it was losing 129% uh, of every revenue uh, dollar they gave it. So the primary attribution, I think that explains this extraordinary efficiency. As you know, Claude loves loops, Boris loves loops, Dario loves loops. And there's two loops going on in Anthropic. The first is the improvement loop. Basically, there is the cycle of you have Claude, Claude leads to better models, what better models lead to more use cases, more use cases generate more feedback, and that leads to better Claude. And then there's also the investment loop. So, revenue from uh, expanding adoption leads to higher growth, higher growth leads to more investment, and more investment leads to more capabilities that leads to better Claude. So, these two loops basically spin together and form an infinite loop that is driving the increase in efficiency until the end of time, according to the Anthropic S1. So, I think that is our transition from the [00:04:01] overall efficiency story to how they execute it. And today we're going to focus a lot on the Claude platform because that is primarily one of the three things that they call the diffusion of Claude in the Anthropic S1. So, they call this Claude platform or in comprehensive development environment. They provide customers with direct access to the primitives to build, deploy, scale their own applications that leverage Claude's capabilities within their own products and workflows. I think one thing that comes up a lot in the discussion in this podcast is the concept of ant fooding where Claude is a lot of Claude's products is actually built with the same Claude platform that is available to you. So, something that's also interesting in the Claude platform that was disclosed is over 50% of their customers with $100,000 or more revenue on the training telephone spaces use Claude platform. And so, that is a increase of an adoption and emergence of Anthropic not [00:05:01] only as a model that, but as also a hyper scalar and a Claude platform. Which I think maybe is in the name inherent in a Claude platform in a Claude platform, but is going from theory to actual practice. And this is something that we're focusing on in the podcast. The other thing that I also really I try to emphasize, which I talked about before, is Claude code, Claude co-work. They all use the same primitives that Claude platform exposes. So, I think that's really interesting. Ultimately, it looks like Anthropic is really positioning their IPO to talk about going from chatbots to co-pilots to virtual collaborators. They use this term instead of agents, maybe it's more friendly, but it's a lot of the same things that you look for from agents, from long-running autonomous work to judgment, self-correction, and iteration, and tool access. So, owning a goal and working alongside of you instead of working inside of one app or task. So, that's ultimately the context I want to provide for this today's podcast. This was recorded when Opus 5.5 [00:06:02] was launched, and I would say overwhelmingly among all the high-taste testers that we talked to, Opus 5.5 is the best model in the world right now. And in topic, it's even just sitting on table 5.5 in probably will release it in the next 2 I don't have to have inside information, but it makes a lot of sense for the IPO. So, let's look forward to that and learn a little bit more about how they built Opus 5.5. >> Yeah, so maybe a distinction that I would make is in a lot of the incidents that people are hearing about and reading about, these are incidents that are happening to models, not just out like everybody's models, right? That are in training. >> [music] >> Where the way that you're running these models is you like put them in a box and you say like solve impossible task, I'm going to stress you out, ready set, go, right? And you [music] like can only do what you can do within your environment and the agent is basically incentivized to try to get out of his box and go solve the problem. When people are running production workloads on our API or on anybody else's API, you're running an environment where like there are safeguards that are running [00:07:03] that are like not necessarily running in training where like the whole point is to actually train the model, [music] right? But in production workloads, first of all, you have those online safeguards, which is really great. Second of all, you have more of this architecture and infrastructure around what's happening. Like the agent isn't just in a box trying to get out, right? Like the harness is over here, the hands are over there, and um it's just like not even a problem that you would have to contemplate. >> [music] >> All right, we're here in the studio the day after Opus 5.5 with Caitlyn and Angela. Welcome. >> Thanks for having us. >> Uh congrats on launching the top model in the world right now. I would say depends who depends who you ask. >> Yeah. >> If I say ask, um, what is the main takeaway personally for you guys? You know, you have you have your official blog posts, the the pricing cuts are really cool. What do you take away when you launch a new model like this? >> We're super proud of this one. Um, the team has put in a ton of different work to just go make this one happen. And I think we've like really heard the [00:08:03] feedback on Opus 5's writing is like kind of convoluted. We really tried to clean that up and I'm really glad to see that like it seems to be resonating with people. Internally when we like tried out the model we felt the same way, so it's just really great to get that validation. To your point around the pricing cuts, there is uh, a lot there and I think increasingly people are looking for good intelligence per dollar. Um, so making sure we get the per dollar part right um, has been really important. There's some stuff in the edges I don't really about like, you know, maybe stuff that we haven't talked about in the blog post, which is that we did do a lot of work on trying to make our API harder to distill from. And so there has been some changes that we have made to go and make that a little bit more concrete and these are things about um, you know, like how do we do like prefix block thinking or how you can replay thinking tokens across different types of models. Um, and there's been a lot of like work and effort on our part around making sure that we try to adjust things for very valid use cases for where you might want to go and do this and like potentially in your compaction strategies and things like that versus in scenarios where um, the use case of [00:09:03] that is actually from like a bad actor effectively. So there has been a lot of threading the needle around those fronts and I'm glad to see a lot of that go out as well. Um, in addition to just like shipping a model that's uh, a good intelligence uh, for for your dollar and also like just communicating a little bit clearer. >> Yeah. Uh, specifically you're talking about uh, this paper stealing reasoning traces from proprietary LLMs. >> Yes, that paper. That one. >> far as, right? They obviously they did a responsible disclosure with you guys. >> responsible disclosure and we're super thankful for for them coming to us >> But could we uh, for those who haven't heard about it? This is really cool they you do it. I mean, it's patched, but like Like what what was it? Like What did it cause a freak out in in in Anthropic? Probably wasn't that big a deal. I don't know. >> I don't know how much this paper specifically caused a freak out because this is something that we knew was a gap. >> Yeah, so there's two separate things. There's the Chinese model distillation, and then this was a specific technique that was published that was high-profile. >> Yeah, exactly. I We We've known for a long time that chain of thought extraction is one of the most powerful [00:10:05] ways that you can distill models. And actually something like how do we preserve thinking blocks across turns and make that something that's more required is something that we've explored back for a while. And what we Where we were at was we always have this really really hard trade-off that we have to make between user experience, like user experience changes that are maybe painful for users of the API, versus protection against distillation. And so, how do we kind of walk that line and say, "We're preventing distillation." And maybe just to take a step back. The reason we don't want distillation is because the entire point of why Anthropic exists is cuz we want AI to create positive outcomes in the world safely. And we run a lot of safeguards classifiers on the tokens that we're producing in order to make sure that they're safe. If someone distills our models and they run those models without safeguards classifiers, there could be bad outcomes in the world, and that's something that we don't want to have happen. So, we don't want people to distill the model specifically for the [00:11:05] sake of um just like using them for malicious reasons or for bad reasons. And so, it's always an interesting trade-off of like for all the people who are using the models for good, all right? Like we don't want to make them have a harder time getting the intelligence out of the models, but we have to make some trade-offs in order to prevent that distillation. And so, we had been working for a while on how do we prevent that chain of thought extraction? Like what do we do about preserving thinking blocks in a way that wouldn't be too painful for people? And we kind of think about user experience hits in like two layers. Like one is just like you have to make an integration change and it's annoying, but you can still accomplish all the same things that you otherwise would have been able to accomplish. It's just a little bit annoying. And then two is like actually you can't accomplish the same things that you could have accomplished previously. And that's the one that we're like really really really trying to not have happen. And so where we landed with the last couple models is a few incremental changes to have like enforce these preserved thinking blocks across turns. And now like starting to be more about [00:12:06] like across, you know, like if you're using multiple accounts and things like this, right? Like how do we make sure that we're um ensuring that you aren't able to take advantage of this chain of thought extraction with the models. And so we shipped a couple additional features to help with that. We worked actually with a lot of open source harness maintainers. And we reached out to them, we talked to them, we were like you're going to have to make these upgrades within your harnesses. We want to make sure you have a good time. As well as obviously our big customers like who are building harnesses on top of the API. And for some of them, a really good example is they have compaction strategies that would have been broken by the need to preserve these thinking blocks. And so we shipped a new um compaction feature within the API to replace what they had otherwise been doing manually, but still works with preserved thinking. So this is an example of like we're walking this line, but we didn't just like kind of YOLO out like now you have to preserve these thinking blocks in this exact way because we would have broken a lot of people, but we were able to take the time and work with customers and open source harness maintainers and give them [00:13:06] what they needed to be able to make that upgrade. >> While we're on the topic of it, anything on the other side of distillation? So uh you know, the other half. >> Yeah, no we we have seen that it is harder to distill like through the measures that we are able to to take a look at. It has been significantly harder because this is uh and you know, credit to the researchers here, they they did disclose it uh very responsibly. >> Yeah. >> Um and so closing it down was great. Um and they did publish the results very publicly. I know it did kind of trend quite a bit on Twitter. So many people could could use it. But yes, we have seen it that actually from the areas that we are able to measure, it does seem to have reduced the effect the negative effects that we don't want. And yeah, the bit I would just add to what Kaylin just said is I think at least I see a lot on Twitter or X or whatever you want to call it. It's a commercialization problem or something like that. You know, if you would have still a model and you can take this and it's you can run it for cheap and and so on and so forth. While true, it's actually not something we really care about very much. We care much much more about this from like a safety point of view more than anything [00:14:06] else. And so that's why we we tend to think about it from that angle more than just like commercially. Realistically, if you distill from us, we you actually pay us. So commercially, we like make the money. So it's not really that kind of an impact. It's just more on like making sure that our model's like we want to make sure we want to make sure the cloud is like good and cloud could be derived. And if it turns out negative, that would be unfortunate. >> Just generally also on 55 cuz you you guys are just fresh off this. One thing that's really fun to establish the dynamic between you two because you also do a lot of appearances together is you're running your head of products, you're head of engineering. How does this play between the the two of you? Especially [laughter] for model launch. Let us take an example. >> So for model launches, so our team works together very closely between product and eng. On my side, a lot more of it is about like what is the balance of trade that we need to make between like launching something that has the right like intelligence bar but also the cost to preservation. Like a lot of like our our updated cash costs and things like that are indications of us being like we want to like encourage you to take integration [00:15:06] changes that are beneficial for actually everybody and just like pass the savings on. But a lot of these kinds of decisions around that and then how do we balance some pieces around our safety motion as well as making sure that it's like an awesome model that people can can take advantage of. And so those kinds of traits around like I would say maybe the qualities and surrounding components around how we launch the model are more on product. And then, from an integration experience standpoint, a lot of that is from like the engineering team directly. So, we work obviously, if you're like talking about the experience of an API, that's like very, very, very tightly integrated. And so, it's actually been really awesome. We have a lot of engineers who actually are deeply product-minded, and we actually like love that. And similarly, on my side, um there's a lot of like product managers who are much more technical. And we want this like blended kind of interdisciplinary and type of people, because then they can kind of analyze it from all the different angles to try to figure out how how uh to do that best, but I tend to lean a little bit more on like commercial impact versus I think [00:16:06] I'd actually frame it as more like the engineers are going to be care much more about the integration experience. Um so, they almost take more of a UX angle to it. Um where the UX is obviously more about the API um and the developer tools. Yeah, it's fun to be a product engineer who works on developer products and platforms. So, I would echo that. >> we have a lot of engineers who love thinking about the product, who love thinking about the user experience, who love thinking about the developer experience. I think it's a dynamic that we have between ourselves, too, where um I have a lot of opinions, and I'm annoying with our product team on how we think about these things, and our product team is also really technical, and they can get in the weeds of the decisions that we're making for scale or for whatever it might be, um which is a fun dynamic that our team has. >> Yeah, something uh Kelly and I have worked together for a long time. >> At Stripe and >> I at Stripe, and then I I I guess we did not work together at OpenAI, um but that's just you. >> labs. >> Um and then I came over, but uh I think some one interesting property of both of us has been Caitlyn frequently actually gets mistaken for a product [00:17:06] manager, and I frequently get mistaken for like someone in engineering, um even though like those are not actually our job functions, um and so, I think that pairing tends to like help out a lot with just making decisions that are consistent across both sides. >> So, I I think I come across a lot of companies with a similar like product leader, engineering leader setup. As software engineering gets solved, maybe those two roles merge. >> I kind of think so. Yeah. I think everyone tends to lean towards >> you know, separate still, then we we're not there yet. >> [laughter] >> We're definitely not there yet. >> So so like why, right? Like what's left? >> I think that like software engineering getting solved is I would I don't know if I would ever use language quite that strong. I think there's that I tend to think about these things like ambitions will continue to elevate, right? And maybe a good example, like a lot of what our team is spending time on right now is, of course, obviously, all the things that we have to do to launch our models safely. It's actually the most important thing that we're doing because of all the work that we have to do on security with agents, with offline safeguards, and online safeguards, all these things that are happening. [00:18:06] But the other half of what we're doing is trying to scale. Um and part of that is because when I joined Anthropic a little over a year ago, we had the messages API. It was a Python monolith. We were running it in a, you know, single GCP region, and everything was fine. And none of my friends and family like knew what Claude was when I was going to take this job, right? And now we're in an extremely different place. And so, the metrics are really wild in terms of how much volume we've seen and have had to support on the API. CPU's really constrained, uh which has been a really interesting one for us. Um and so we've done a lot of work on how do we scale these systems? And that takes a lot of obviously just like distributed systems expertise and certain language and frameworks expertise that the models have, but when it comes to operating and evolving a system at scale like this, we're not quite we're not nearly in a place where it's like good luck, Claude. Have fun. And just go and kind of make all the changes you need to make to the system, roll them out safely, and operate this thing at scale. Um there's still just a lot of work for humans to have to do that I [00:19:06] imagine will continue for a while. >> I'm curious how that's changed. So like a while ago, you know, when Fable first came out, we had Glasswing and there was a whole delay and change in how models started to get released. We saw this for most Frontier models. Uh there's safety testing. What's the status of things now? How do you guys see it? How do you determine like one thing is there's a new Sonnet and um Haiku also coming. I assume those will just come out, but for something like Opus 5.5 or a new Fable, what's the what's the current status? >> Yeah, so I would say this starts with what our kind of policy decisions need to be on safeguards and on safety. Um and that's both offline safety and online safety. And so there's been a lot talked about lately around data retention and what we're doing there in order to what you know, release these models safely. And the reason for that is because you have, you know, situations where there are many, many accounts across like a whole giant network of people who are choosing to spin up accounts and use the models in [00:20:07] various ways. And there's no way that in online way as tokens are flowing through the system, we would be able to kind of catch that there's something malicious happening with the models or distillation or whatever might be happening. And so we've started to have to have data retention policies so that we could do automated review and human review whatever it needs to be to make sure that good things are happening with the models. And with each model that we're releasing, like that's one example of a policy decision decision that needs to happen on like what will the policy be for this specific model. Um and so Opus 5.5 that launched yesterday, for example, does not have those data retention requirements that Fable has, right? But that's a decision that we have to make each time we're launching one of these models. And if it is going to have those policies, we're going to have to do the work to make sure that those things are enforced. Similarly with programs that we have, like we have cyber verification programs and now newly our life science verification program which says, you know, like you have a difference in how the online safeguards are working for the model [00:21:08] that you're using, making sure all those programs are in place is another one. And so, it's just kind of a flavor of the type of work that we've done as part of each model launch that these are everything that I'm saying right now are not things that we used to have to think about when we were launching these [laughter] models. Um and now it's it's the majority of the work is like, what are we doing to prevent distillation, the online safeguards, the offline safeguards, the policies. A really great example is we built a system to make sure that authorization for being allowed to use a model or being allowed to use one of these like programs that give you maybe like relaxed safeguards. That authorization threads all the way down through our system in a way that we can really trust. You know, it's not like, oh, you got access to Mythos and so we throw you an API key and we flip a feature flag and call it a day. We have a really robust system on how we're doing off and making sure that if you're supposed to be using that model, like only you are using that model in the way that you're meant to be using it. So, there's a lot of complex things that are starting to happen around model launches that, again, we just never used to think about them. >> Yeah, and it's it's scaling up with [00:22:08] capability. Like as our capabilities get more and more advanced and they start to spike, you know, in specific areas, that's where we need to pay like extra attention. And so, a lot of that evaluation comes just from like those very particular capabilities that we might either see as like offense dominant or defense dominant and therefore might take different strategies for all of that. And as we bake that all the way through to eventually the model launch, there will be components kind of like Caitlyn mentioned where we kind of don't exactly know how people might use those capabilities. There's some more obvious ones like between coding, right? Like great, go make awesome products, go ship more things. Um it can be really awesome for you. And then other areas like bio and cyber, it's like it could be dual use and there could be negative effects, but obviously of course you could be a really valid cybersecurity researcher, you are trying to find vulnerabilities, you are trying to be a good actor. Similarly with bio, you might be researching and stuff like that and so a lot of these programs like the one that you you have up here are to basically provide access to that dual use. You do have to specify your use case and things like that. Um but we do want to enable the community [00:23:08] uh to be able to take advantage of these capabilities and see what amazing things that they can do with it as well. >> Well, we'll put up the links and I maybe we'll dive into them later. There's also one for bio. Also, I just think at a high level, you know, it's it's really important that you guys actually think about this stuff. People forget like Fable was rolled back and there's all this distillation and people are like, "Ah, zero data retention is not a thing on Fable. Crazy." But you have to remember like, you know, it got pulled back. Uh there's distillation attacks. Like it's good you're overthinking and doing it in the right way while releasing it, right? >> Yeah, and a lot of that is like building the right framework um just so again kind of talking about the end side of it, building the right framework in our system to be able to do that. So, I talked a little bit about this like this idea of like authorization and trust grants that we've built throughout our system now. >> Never seen ever, right? Not Not this Not this precedent. >> Yeah. >> Uh but also it's like a very common thing in platform engineering that you like uh you have one incident and then you like over prepare for that incident and then the incident never happens again. >> Yeah. >> [laughter] >> Which is I feel like I should have said [00:24:08] like, you know, now you have the button. >> Well, it's true and it's funny you say that because it was a like with everything going on, we then had this line item in our you know, the work that our team is doing that was we're going to put this system in place that makes it so that it could be a single button to pull back for everybody. But um and so it's funny cuz it's like, do we justify that we're doing this work cuz we don't we hope that something like that never happens again. But I think in this case, it's really just a laddering up of those same primitives that we wanted to put in place everywhere to help us manage these verification programs and access to models like Mythos and otherwise to just say like, just set that same thing below like everybody's model access so that in case we ever needed to, we had the right controls in place. >> Yeah, and decisions like this have actually proven to be like we'll see about this one in particular, but examples like this have been consistently the way that we've had to actually transform the way we develop our roadmap because, you know, in traditional product management, I guess, you can be like, "Okay, like these are the set of things I hypothesize to happen. They will have like a certain sequence, a certain ordering, but you are picking bets in a way that are much [00:25:09] more concrete." Versus like, you know, the D deployment of Fable, for example, I don't think anyone of us would have guessed that that would have been like a valid use case that would have happened by the government and the way you just like had to go do it. But clearly, the probability is non-zero. And so, from this form of thinking, it's become deeply probabilistic, actually. From like a product management standpoint of like, "So, should we prioritize this? What are the chances of it ever happening again?" When we look at things like that, we're like, "Okay, if we are to imagine the most crazy set of possibilities, we probably actually have very little certainty as to where things will go." And so, we actually just kind of need to like hedge and build a portfolio of bets. And so, more often than not, we will take hedging scenarios of like, "Okay, we don't think it would happen again, but in the event that it does, we would actually want that hedge or have the ability to be able to do that very well." And so, we will prioritize things like that. And then we have increasingly found that more often than not, by taking that strategy of product development, most of the time, we actually find ourselves being like, "We actually do use the [00:26:09] thing." >> Is there an another example that's a separate one just so I can understand? >> For example, this this access program. This access program is a really good one. This It's internally it's it has a bunch of granting and all these kinds of things. We developed it specifically for reasons about like being able to actually give programs like this out to customers. Like very very specifically. >> what the access program specifically is? >> Sure. So, if you want dual use if you want to So, for let's let's take a model. So, you take a model, call it like Opus, and you want to be able to use advanced cyber capabilities on on Opus for research or for whatever you want to go do. Like you're a you're a defensive user. You can apply to this program to be able to with your use case. And if approved, what we'll do is actually diminish the safeguards on your version of the model so that you can then go ahead and try to take advantage of this like cyber activities for for your use case. Now, obviously that requires like per use case understanding. It requires scoping down to a particular user cuz you wouldn't do that generally. You'd have to like apply. And even things like this like [00:27:10] this kind of program, we designed a lot of the primitives that Caitlyn was talking about to be able to handle things like this. Now, it was surprising to us or maybe like you know in the fully approach, we found that we were actually able to take this and make it ex- We were like when we designing it, we were like what kind of should it be? How generic should it be? Should it not be too generic? Should it be overly focused on making this verification, you know, style portal plus granting of a very specific capabilities of the model. And we tended to opt for like more genericism because we were like we don't know. Like maybe we can use this for other things, maybe not. And then we ended up being massively useful for things like you know, access to the model writ large for managing any of our trusted access programs for handling data deployments of models because we were actually very very generic about how we handled every single primitive. From a normal product standpoint, I would normally probably be like that probably doesn't make a ton of sense. We should focus on the use cases that make sense. But here increasingly because we choose the more generic approach, we find ourselves reaching for that solution more often than not and it pays like [00:28:10] increasing dividends for that reason. >> It's just something that you needed to have anyway and you just had to force a function to have it. On a technical basis, what so I'm I'm from the AWS world. I I I am so I see and then people like who get really tense about this talk about Zanzibar and all these all these other technical architectures. I'm not familiar with the word token grants or or like trust grants. Trust grants. Trust grants. Is there a difference between normal off C and and what trust grants are? >> I mean not really. It's It is It's a It's a form of authorization. Yeah, kind of. I think so really we have like Um, like authorization of the form of like, okay, you're you made a request to an API that's like under the hood is going to access some data making sure it's like your data that you're accessing, right? Like really, really vanilla authorization. This is just like we put a data model in place that's like a very clean like a trust grant means like you have been given access to some specific model or you've been given access to some specific program or you [00:29:10] know, like there's a a bunch of different permutations of this. We actually find ourselves reaching for it all of the time. >> Mhm. >> Um, another good one was we first rolled out um, some of this preserve thinking in uh, or for the for Fable 51, we kind of grandfathered in some people and gave a bit of a, you know, just like a time yeah, grace period to go and uh, make the integration changes. And as part of grandfathering in, we said, okay, we have to like track who is like an existing customer who needs the extra time, right? And then we did a trust grant and so then we're able to say for anything, we're just running the service that is kind of on the hot path of our request to make sure that um, you know, we just like flow through and and check whether you have the grant for the thing that you're trying to do. Um, which is it's really not any different than like vanilla authorization, really, but like we just had to spin up a system that was specific to our system and like where we needed to be able to insert these checks. >> Yeah. I mean, I I think uh, these are all like the engineering [00:30:10] things that people kind of take for granted, don't don't really think about. What is another example, I guess, of the platform stuff that had to happen, you know, maybe serving cloud code, uh, maybe serving, you know, obviously I talk a lot with Cognition, but other customers just in terms of a model launch, right? Like we'll talk more about the uh, the the the safety work that is that's being happened. I just wanted to sort of go at a high level because it's fresh off your memory, like what goes into a model launch? >> [laughter] >> It's become a bit of a a machine these days. Uh, minus the safety stuff, which I think is is the more complicated part, but outside of those pieces, there is a lot of getting feedback from our customers. So, you know, we make the model, there's a lot of safety testing, there's also evaluation of where we think it needs to be, we make the required trade-offs and decisions as necessary, and then we have to go through testing. We'll cut various versions of the model, we'll evaluate it, we'll do a lot of internal and footing is what we'll call it, and that actually always really helps. Like we catch like really early things like guys, this is, you know, maybe not this, [00:31:10] and we obviously use Claude a lot for our work internally, so we're able to actually see early signals around stuff, and it helps us detect like, you know, which which variants might be better than others. And then when we feel really confident with like a certain set of variants, we will cut a version for our for our customers. Sometimes that will go through our first-party products like through Claude Code or Co-worker or things like that. In other cases, it will go through our API customers, and we'll engage in a kind of like a early access program effectively with those customers. They'll need to make changes into their harness, and we've engaged with a lot of customers like like Cognition, like like Character, like these folks, and we have a bit of a process on how we handle that so that we all kind of get they all we get the feedback that we we need. Like does it perform well in your e-vals? Do you like it? What don't you like? Some things are are vibey, and we like to hear those things, too. And some things are just like straight-up e-vals, like did it do well on the things that you care about? And we take that all together, and once we all feel good about like all those different pieces, we launch the model. >> The state of model feedback is feels very unstructured still. Basically, it's always a Slack channel between companies [00:32:11] of like, "Hey, this thing seems weird." And then people are like, "Okay, I'll fix it, I guess. Give me an idea or something." >> Yeah, it's very I mean, it's it's the nature of like a product that's like non-deterministic, right? Like, but I think to your point, the only customers really that are part of our early access programs are customers that have really robust e-vals. Like I do think a lot of this is vibes-based, and I think a lot of the conversations tend to be kind of like vibes-based and and on the fly. And part of that is because we're like racing, right? It's like we want to minimize as much as possible the window between we have a version of the model that we've cut for testing and we give everyone access to that model. And so it is just like this really tight window, but the customers we tend to do with are the ones who like we know they're going to go like kick off their whole eval suite and be able to come back to us with like really actionable feedback as quickly as possible alongside the vibes. >> Um. >> The vibes based feedback, so um of course there's like definitely like processy things that we could do to make these things better over time, but I [00:33:12] mean you'd be surprised I mean you probably wouldn't be surprised, but I think some people would be surprised about how few, you know, like products that exist in the world that are agentic that like don't have good evals to to be able to help test these models. >> Boris on the pod 2 years ago saying we don't have evals for Claude code. I mean that's changed, but that's changed that's changed, but you know, like yeah, a lot of things like you know, your evals are your users and you don't have time to formalize them or you don't have the experience or whatever and that's fine. At some point when you get serious you'll you'll have evals. >> Yeah, totally exactly. >> I think the good thing is you guys do a lot of amp fooding as you say, so it's commonly said that you find a lot of this testing it yourself before it comes out, right? >> Yeah. >> Yeah, the APIs customers in particular are very useful because they'll put Claude into a different harness. Um and we want Claude to be very very general. Uh so it's ability to work in other harnesses is quite important to us. And so that style of feedback tends to be exceptionally useful for us. Because it's it's novel um from what we would have gotten from amp fooding. >> What kind of feedback helps? Is it like regressions? We have regressions here. [00:34:12] We notice good attributes here. >> Yeah. >> Uh the good stuff is always it's always very validating. Uh I think we're very frequently not surprised, but we like to look to be validated on some of those things that we might hypotheses. Honestly, the most useful feedback is negative. Um like you know, it didn't do well at this and do well at that. We love that feedback and then we're like great, how do we fix it? >> Yeah, and it tends to some of these things that are like it's really hard because not everyone actually wants like Claude to land at the exact same spot on every spectrum. And and maybe a good example is like you know, laziness versus proactivity and taking action. And so there's sometimes when like we have a model and and people are like like just finish my task. Like stop being lazy, like do better. And other times when it's like, hey, we were like just kind of feeling this out and planning and then you like went and wrote a bunch of code or you went and like did a whole bunch of action that I wasn't ready for. And getting that exactly right is literally impossible because everybody actually wants something different in reality. But I do think we've had situations where we've cut an early version of the model. We [00:35:12] found that we we saw too much laziness, right? And like we like put it back in the oven for a bit, right? And and tried to train that back out. And so those are the sorts of things that we may not have caught if all we're doing is internally ant fooding within our products and within our API. It's way, way more important for us to get that out there in the API and get our all of our other customers running with it and let us know what we need to change. >> Yeah, there's a core question. I think it's probably the if I if I were in your shoes, like one of the hardest things about PMing for a model, which is let's call it alignment versus personalization. There's a default personality, default behavior that you want for all consumers. And then you want to be able to respond very quickly to someone's Claude MD saying, no, no, no, like please be more proactive than default. Uh and then like so as to my knowledge, you know, and I was involved in Frontier Code, uh we were involved in the Fable launch, uh which put Frontier Code on the map. Uh and now Opus 5.5, we have it back on on on the the model [00:36:12] benchmark. That is a very big part of Frontier Code, right? The reason that the model curve looks weird for Frontier Code is because it penalizes too much activity or too much uh files. Uh which is good. Like that was a whole design purpose of Frontier Code. But like I could easily just be like in my cloud MD just say like don't do that and then it won't do it. >> Mhm. >> And so like am I penalizing cloud opus 55 unfairly because my benchmark doesn't have that personalization element which is subjective. So I don't know what the question is but like how do you deal with like personalization? All right, like like clearly a model can do so many different things and it's so different everyone has their own cloud MD. Uh how do you evaluate that? >> Yeah. We don't do a great job of it. There's like there's seriously it's it's really hard. It's super hard and there there competing and it's Yeah, yeah. >> I thought I thought there's the cloud constitution I think. >> That's old. >> There there is a constitution but I'll [00:37:13] I'll let you guys answer, you know. >> The constitution's helpful but but for things like this on how it personalizes to instructions which are different and preferences which are very very different across these things is very hard for us to do. We generally try to solve this by having more parameters that then allows you to tweak it per customer base. And if we don't control your customer base and you're an API customer then we try to give you recommendations and then you try to do that tweaking. So there are things like you know around it doesn't exactly it's not exactly at the same level like personalization to the cloud you know various cloud MDs or now agents MDs or whatever but we do try to give things around the layer of like level of effort like how much effort you should put in like what are the like the level like maybe the size of the model that you might actually like try to care about things like that but it's I think it's a it's not quite there for what we would actually need for like that level of personalization. So it's a very hard problem and right now a lot of our way to solve that is beyond the primary like the giving multiple parameters to something is try to exhibit conflicting taste honestly [00:38:15] across like all of the feedback that we get and try to try to make like a judgment call on some of these things. I think over time they actually and as you scale to more customers and more users that's like not a very scalable way to go do something and we will need to develop ways for Claude to kind of like take in your personalization and maybe that's like a signal you can provide or something you could tune on or things like that but I think we'll have to like take a route that's much more oriented towards that direction but for now yeah I know we we actually just try to make judgment calls on a lot of the stuff but it's it's pretty tough. >> [laughter] >> It is what it is. >> A small a small note there I guess the default effort reasoning changed it's now medium. This is one of those things that probably affects the average customer that doesn't use this toggle. >> Yeah I say it's a Claude code setting right? >> It's a Claude code setting the default is to medium. >> Yeah they just >> And and this and this thing varies this thing does like this is a this is a perfect example of a parameter that we use to kind of like personalize I I would say this is maybe a little too aggressive of a term but we kind of like customize maybe for different sub segments so like the default setting will adjust for whether you're a free [00:39:17] customer or prosumer customer or Claude Kid versus coworker versus enterprise like things like that versus API and stuff like that knowing that you can obviously always adjust it but those are kind of like light ways where we try to give that. One thing that we are hesitant to do is we don't want people to feel like they can't control their customization of Claude. So we could take a heavier hand on these things for example but we want to give users the ability to experience the intelligence levels of Claude as they kind of see fit especially if you want to go towards like the higher end or you want to experiment with maybe sometimes in the in the lower end if that actually is like faster for your use case which is a hard product trade-off I think and so we we tend to lean towards gentle defaults but give you options to to adjust which I think works okayish for now but I imagine us you know like learning how to do that better over time for customers. We're not like super super happy with any of this yet but we know as we launch a model like you asked like what does it take to launch a model? One of the big things obviously that we have to do is like okay Claude like got good at a thing and like [00:40:17] sometimes that thing is something we should expose as a parameter on the API to let you control. But I think as this has evolved over time, we've ended up with like this funny mishmash of like there's effort, there's budgets, there's all these things that you can set on the API and it's like how can anyone like possibly figure out what is the right combination of things for any given task for what you should go and set? And it's something that this is us just very much like figuring it out as we go and figuring it out with customers and figuring out like which of those parameters do we deprecate over time, which of those parameters do we just like default better, which of those parameters should be like a 1 to 100 scale versus a low, medium, high scale, right? Like because actually we've trained Claude to get really good at understanding like a whole spectrum of options on some of those sorts of things, but like do we expose the like entirety of those millions and millions of permutations of options? Like you wouldn't really get very far and that's one of the things actually that um is part of the value prop with something like Claude managed agent, which is like still programmatic access [00:41:18] to Claude in an API, but you get a harness out of the box where we do every time we're launching the model a lot of the work to go figure out like what are the new parameters, what are the breaking changes, how do we get the best out of Claude within our harness so that for the people who are using it, um you just kind of get that out of the box, you don't have to think about it. >> Just for people who are uh drawn to that and which there is a significant and growing population, particularly we actually did a whole CMA workshop at at EIE which I thought uh was really well done. Is there a separate feed for people to subscribe to like updates on CMA? >> Feed like >> Like the Twitter feed, newsletter, I don't know. >> No, I think actually our our Claude dev channel we are yeah, we're we're pretty loud about it um because it is like an interesting option for people. Yeah. >> Yeah. I just want to give people like if that's your if that's your way of consuming Claude, that's the this is the path. >> I think it's also pretty underrated. Um for those that aren't consuming Claude this way, what is Claude managed agents and why should people use it? >> Oh yeah, we should probably do a do a [00:42:18] little intro. I don't know. That's something you house. >> Yeah. >> Yeah, yeah, yeah, little two two sentence thing. >> I think a lot of people glanced over it when it came out, right? Everyone was in this phase of I need to do harness engineering, I need to control this. And like you hear all that and now it's like do you need to do all that or do you just use managed agents and you have a lot of infra you can skip over setting up but >> Give give the quick three. >> Yeah, it it depends on your use case but so Claude managed agents is our API today. We have the messages API which is essentially raw token access. Then we have Claude managed agents which is a suite of APIs but it lets you define an agent, start a session with that agent and it's essentially just kind of a higher order abstraction that lets you skip setting up all the infrastructure, setting up your harness and doing a lot of that type of work. It's a much more out of the box integration for people who want to programmatically run agents. >> Does CMA share infrastructure with Claude tag? Because the sequencing of it seemed very like oh [00:43:19] like that's the >> Yeah, yes. >> Claude tag >> A lot of it. A lot of it. So one of the big things that we do is as the platform team we run our APIs, we also kind of run all of our foundational product infrastructure and product services. So like really good example is like our team runs our sandboxing fleet that backs both Claude managed agents as well as products like Claude tag and otherwise but at the actual harness layer for those two products like there's just differences in what we might want to do within the harness so that they don't they don't actually share a harness. But they share a lot of the infrastructure, the powers, the execution and powers the APIs. >> I think that's that's really people should use this more. I think everyone basically had a copycat clones of managed agents after you guys launched it which you know imitation is a sincerest form of flattery. Since we're on this topic, we'll briefly mention uh all the memory stuff that that's been going on. I feel like it was a big part of Anthropic's work in the first half of the year. Has it stalled out? You know, is is it still still ongoing? Like, [00:44:20] what's what's the emphasis here now? >> Managed agents specifically? >> Memory, dreaming. Dreaming, >> Um it is a big part of what we're doing. I think maybe the way to think about it is we launched managed agents as a platform for people who like didn't want to do all these things themselves, right? Or just wanted to like offload like I'm getting the best out of Claude because our team is literally sitting here like evaling all these things and making sure that the harness is going to be really powerful. Memory and dreaming is all about just like how do you make sure your agent can continuously improve over time. And all of that is still an important focus for our team, but what I would actually say is to the point of what we talked about earlier, as launching models has become harder and harder from a safety perspective, I think most of our team's energy is going less into like, "Here's a fun feature to like hill climb on this agent or whatever." and much more into how do we bring safety into all these things because the most important thing that we can do is get the power of the model out there and get it out there safely. And so, I would say a lot of our energy is [00:45:20] take all these same concepts and give you an agent system that's going to be really powerful, but also that's going to be really safe. And so, that's where we're putting our energy. >> I think compared to maybe earlier this year, we have slowed down on like skills, dreaming, memory, like these kind of like moment in time injections. >> skills. >> No, no, no, we love skills. We love skills. Um we we are making we are making updates. I'm not saying that we're not doing anything, but um we we love skills as well. But yes, I think we've slowed down a little bit comparatively um for some of the other efforts. But uh we are very bullish on uh people kind of doing some level of hill climbing at the harness layer. Uh to, you know, some of the previous conversation here. We would do a lot of this ourselves, but the harness that we provide is very simple in a way. It's It's intentionally simple. Uh we don't want to be actually overly opinionated. Want to make sure the best practices get merged in and there's some kind of undifferentiated pieces of the harness uh agentic loop that yeah, are uh hard to string together and also like you can kind of just boilerplate do it. Now, the interesting part on top of that though [00:46:20] is like once you can get that base layer, there are these like hill climbing strategies that we construct. And like Dreaming is actually one of them. So, Dreaming under the hood is a multi-agent system that we hill climb on top of CMA and it's actually a CMA on top of itself, basically. >> CMA on top of itself? >> It's a It like spins up another Yeah, uh the way Dreaming works is that we spin up another managed agent that then can inspect the previous managed agent and then write uh memories to that managed agent. So, it's managed agents all the way down. And this is a bit of a thing by design because we envision that you could probably put together a lot of these modular components for something interesting. And Dreaming is just like an example one of the more interesting things that we were going after. Uh but we kind of have been looking at this and it's being like there's actually a ton of other stuff you can do. You can do like best of N as a type of like multi-agent strategy. You can do a verifier agent uh that kind of like goal thing or outcome thing. Um there's a bunch of other stuff in there and we've seen people do some pretty funky crazy stuff. Like there's been a harness that like shares like crazy memory across everything and they spin up like tons of parallel agents. And we're like we should actually just probably enable you to do that. Uh in [00:47:20] the same way that we did we built Dreaming that's like managed agents all the way down. There's a update that we'll make relatively soon um where we're updating the underlying primitive. We'll expose it uh to allow users to kind of like construct harness strategies maybe or however you want to call this stuff. >> strategies. >> Multi-agent strategies that allow you to kind of build your own version of this. And we'll give you like, you know, templates of like this is how we started with Dreaming. Maybe there's like an update to this that you would want to make as a developer and you would find some more value out of it. >> Yeah, I think that's a really good framing. It's less that we're slowing down on this sort of stuff and more saying like we've got the primitives in place. We can continue to let people just like take and run with those things while giving them the tools to say, here's some like cool multi-agent strategies and like things that we've put into into Cloud Manager just to make it really powerful. Now you can like what Claude do work to take advantage of those primitives and make it even more powerful yourself, like create new multi-agent strategies and whatever it might be. While a lot of our team's attention turns to, okay great like community run with that, right? And and [00:48:21] you do you while we continue to focus on what are all the things that we need to do to make sure that you get the access to the models within that system safely. >> One thing you brought up was you run a lot of evals on this stuff. >> Mhm. >> Pretty hot topic right now. Any advice, best practices? >> For evals specifically? >> Evals on harness as agents. >> There's a lot of different types of evals, right? Like not all evals are made equal for different, you know, like situations that you're in and things that you're trying to test for. You obviously have just like static deterministic stuff and then you have LLMs as a judge. You have all these different strategies. But I guess I would just say like what we found from most people is like you probably need more rather than less. >> [laughter] >> Um it's just uh there's not like one perfect like couple sets of evals that you could put together that are going to get you all the answers that you need to make your system better. We found that people end up needing like a relatively large suite of these things and then it's like you can have your agents work with you to evolve your eval suite over time as well. Um but I think it's just [00:49:22] like under investing is maybe like the biggest mistake that we've seen from people on that front. >> Yeah, and we start simple. I think some people like go a little crazy and they're like, you know, they like look at all the the menu of options that you kind of listed and you're like, wow, this is so crazy. Let me try this and that and what's the best one? And like while they do all those things like the number of evals they actually have are like zero. So you might as well just like do something simple. It will help. Um >> Yeah, I think some some part of people are inexperienced, they're not trained to do this. Um and they don't know what good looks like. So I think you guys publishing more examples of that was is really helpful. Something I I I do tell people a lot that I'm doing with more Fable and Astro class type models is we're treating engineering like research. And when you treat your engineering like research, you naturally do more evals. You do more labeling. You do more scaling of runs prototyping and scaling of runs. And you're not training a model, you're building the harness, the product, whatever it is. But it you're approaching it like research. You're defining up front what your [00:50:22] goal is, trying to formalize it as an eval, and then trying to identify, okay, why is it failing? And sometimes your eval is wrong. >> Totally. >> Actually, a lot of times your eval is wrong. Sometimes your environment's wrong. And then sometimes the model's wrong. >> [laughter] >> And actually that is like a lot of part what I think AI engineering is becoming, which is doing all this stuff. >> Completely agree. Yeah. And we'll continue to launch more features within Claude managed agents to give people more tooling to do this sort of stuff out of the box. >> Yeah. >> And you'll probably find from us that you'll get something that's like, here's a product that will help you create evals for your managed agents or whatever it might be, and it'll feel very much like working with Claude in order to do that. >> Yes. >> Right. >> That is [snorts] That is the best way to do it. Claude's actually really good at creating evals to evaluate agents. >> Yeah, exactly. >> Actually, because like people like blur blank out when they look at like JSON Ls. They just >> Yeah. >> [laughter] >> Right? But >> Yeah. >> a pretty eval UI where you're like left right left right left right, and you're you're sort of like doing what the human grader data people labelers would do anyway. >> yes. [00:51:22] >> Uh, you can do that yourself. >> You can. >> Just like throw Fable at it. >> Yeah. >> And it'll materialize, but also write comments. So cuz it just like outputs all these ratings, right? It'll give you an ELO, give you whatever. But uh, Claude can just consume that and then work on itself, right? That's the feedback it needs. And people just don't know they can prompt for an eval. >> [laughter] >> Yeah. Yeah. I love that you said that. I think it's like a very like AGI pilled way to like do it. Uh, because it also allows you to update and edit that system like very systematically. And so as new models improve, as things change, you can really inspect where your system is and it'll just like >> That's in every part. >> Exactly. >> Uh yeah, so basically I think it's you know, how we talked about how maybe product and engineering are merging. It's also there's like application layer engineering and and sort of model research are sort of merging in in those techniques that they're being shared, right? Like >> That's very true. >> Model people are very very used to doing evals. Application people know like the the only thing you care about is does it fit fit the PRD? And like >> [laughter] >> Yeah, it's a it's like a whole different [00:52:23] >> It's a whole new skill set, I think. >> I think you guys put out a pretty cool blog post on this today. It's somewhat different. It's about making the UI a lot faster, but to your point of treating it like research, basically define a goal of we want stuff to load faster and then define and you know, question things what can we do instead of wall clock timing. There's fun quotes in here. I think it's nice to see how the actual researchers or engineers prompted Claude. There was stuff like you know, we had we had 12 things. We said you have 2 weeks. It did nine out of 12 of them in like three days or something, but fun blog post for people to research for people to read through. >> Yeah, I love it when we share our actual prompts and and like for some reason people are like, oh god, there was some crazy unique ways of prompting and it's just like try harder, Claude. Thank you. >> Uh though my only feedback is it should link to the workbench, which is no longer a thing, but I used to >> That is that is a good That's a good idea. Okay. Good feedback heard. Love it. Yeah. >> Uh yeah, but you were going to share [00:53:24] more details behind the engineering story or the the rust part that you guys did. >> So I think I talked a little bit earlier about like scale is a problem for us. Um it's a problem for everybody. Or just like I don't know computes at the like GPU accelerator layer and much less so at the CPU you know, like old school layer and that's it's it's just like as much of a problem today as the accelerator layer, which is really interesting. >> I think you you cannot get enough VMs. You cannot get enough sandboxes. >> Yeah, exactly. >> Everything being able to spin up spin up and so >> With all the resources that you have you you >> Yeah, I know. >> You got a lot of customers too. >> Yeah, yeah, yeah. Which is crazy. Yeah, and and think about it because we're using CPU across obviously just like running our product services, running the API, running all of our apps. We're using CPU on sandboxing, all the sandboxing that we're doing and then [00:54:24] just like all the things like CI, you know, like you'd be surprised how much more code is getting written. Therefore there's so many more CI runs. Therefore you need more CPU to handle that. And so on the whole and this is not just us is everybody that I talk to. >> I would add training. Training is probably using a different bucket. >> So there's there's a different bucket, right? And I'm sure they don't comingle that much. >> Well, they don't comingle although like CPU for Anthropic as a whole is is is like a zero sum game, you know, like we get our we negotiate our deals with you know, the CSPs and we get our CPU and we have to allocate within Anthropic how much of that goes to all the things I just mentioned including like research sandboxing versus product sandboxing and otherwise. And so we really are allocating across absolutely everything. And it's just everyone in the industry right now is uh >> Short on everything. >> Yeah, you used to be able to just like you know, like scale up on demand into CPU capacity if as long as your budget allowed you to, right? And like now it becomes something that you have to [00:55:25] capacity plan for in advance, which is you know, like kind of punching everyone in the face. It's brand new and people are starting to do it. And so dealing with that means a few things. Like one is obviously just go try to get more CPU. Cool, we're doing that as much as we can. But two is we want to be able to at the infrastructure layer run it more efficiently, right? Like figure out how we can make sure that we're running our sandboxes as as efficiently as we can without compromising on things like tenant isolation or properties that we like, you know, care about for security or otherwise. But then at the application layer, how can you how can we run our applications and our APIs more efficiently? The thing that I'm like most excited about as of last week we fully rolled out the messages API written in Rust, which is one of those like insane feats of like what you're able to do in the agentic coding era that used to be like we started talking about this like earlier this year we were like we're going to hit a CPU wall, there's stuff we need to do and and [00:56:25] we've done it, right? Like we've gone multi-region, we've done all these things we need to do. But we were like we're running this Python monolith and it's like it's like very CPU inefficient. And so what are we going to do to make the system more efficient? >> Yeah. >> [laughter] >> So we chose Rust, which was a crazy decision at the time because we had a team of people like on paper it's obvious, but our our team was like we don't have a single Yeah, we don't have a single Rust developer in the room. Exactly. >> [laughter] >> Literally. And and not only that, but yes, Claude is going to write the code. The humans like design the system, right? So like the humans are very involved in making sure at the end of the day it's just like a distributed systems architecture, right? Like and Claude can help you both with actually writing the code, but also helping those engineers like you know, bounce around ideas and learn what they need to learn to design that system. But what's most interesting is the two things that mattered most were the plan to do the thing and then the verification of the thing. Cuz the actual implementation was like [00:57:25] Claude writing code. And so what we ended up doing was we literally built like an entire fork of the entire API in Rust and then started running shadow traffic through it and verifying that the outputs that we were getting from the Python system and the Rust system were >> This is a feature flags all you need. >> Yeah, literally. Um and I just it's just I don't know. I never in my career would have had a team say like, "Yeah, yeah, cool. We're just going to like fork the entire thing." people now. Uh well, they would probably call themselves Rust people now. >> [laughter] >> I mean that's a pretty big step to Rust. >> Yeah, exactly. Um and along the way we have hired a couple of people who have more experience in Rust, but yeah, but it it still is just like you never would have signed up for that. It would have taken years and years and years. Like you wouldn't do it all in one shot. You would have carved out pieces of the system and say like migrate this, migrate that, migrate that. >> a multi-year effort. >> Yeah, but you would never like kind of one in one swoop, right? Like port the whole thing over and then just like slowly turn it on, but that is what we [00:58:25] ended up doing. Um and it worked. And so not only do we actually save a lot of CPU, but we uh our latency has uh seen a nice improvement, which is pretty cool, too. Um and you know, it's just it's it's a crazy era we live in cuz you just never could have done that sort of thing previously. >> Uh order of magnitude of savings on the on the CPU side? >> Uh >> Just for people considering this. >> Yeah, yeah, I'm uh okay, order of magnitude. That's really interesting. So >> ballpark, yeah. >> Yeah, ballpark. I guess I would say like we've saved somewhere like 40 to 50% something like really crazy. >> Yeah. >> And then latency I think we saw somewhere between like 10 and 20% um improvement on like time to first token specifically. So a lot of just really cool improvements uh across higher on like uh longer running workloads, but so it'll it'll roll out um soon. Yeah, but it'll get faster. >> Do you fight the the urge? You know, I've been involved in some major rewrites. There's always like, "What if we add this new thing that didn't exist before in in rewrite? Since we're we're [00:59:25] writing anyway, right?" Uh but then like that adds scope. >> Yeah. Yeah. >> So, do you just like avoid that? Is it a pure rewrite? >> It was a pure It was um maybe the thing that I would call it is like a port. What we did was we we did exactly that. We like resisted the urge on all these things. Then uh and because what the team already had to keep up with was a little like SWAT team that we put together to go and do this was doing this while the whole rest of our team was like building features into the API, right? And they did all the like static analysis stuff that would yell at you because you built a feature over here and you didn't consider rest over there, right? And like they dealt with all of that, but they resisted the urge to add features themselves or like make the system different. Now, their next goal is to go and do some sort of re-architecture type work, including stuff that I think will be more user-facing. Like we're starting to roll out some things that are like, you know, what if we were able to not have you have to like pass back and forth the entire like conversation state, right? And um if you didn't have to do that on the messages API because [01:00:26] you were open to like, you know, a storage of that state for some period of time, right? Then like especially for long conversations where the context really builds up, like you get a lot of really great latency and efficiency wins there, too. Like these are the sorts of things that the team that was doing that port was like, "Ooh, like what if we do this, right?" >> Doesn't your caching API have that? >> So, we had prompt caching. Yeah, but you still have to like pass back the state of the conversation. >> Okay. Very, very niche small subsets, but I guess this is the levels of this. >> this is important. Yeah, yeah, yeah. There's a lot of savings that can be done, but it also turns them into a cloud. >> Yeah. And then from from type of >> [laughter] >> from number of users' point of view, it's small, but if you look at the actual token volume on those those types of workloads, it's like going to be the majority. >> Well, I'm very like I What Cloudflare did with Durable Objects, I I feel like every cloud provider should clone. Do you Do you know what I mean? >> Yeah. >> Like right? Isn't that the right shape for a messages API that needs to continuation? >> Yeah, I mean, it is really interesting cuz I think what we're we are experimenting with various like what is the right shape of like how we should [01:01:27] run our compute in order to power all these sorts of things. Um, but it's also interesting because we can't necessarily rely on like because we get our compute across we run on all three clouds, we're running all over the place. Like we don't necessarily have like one shape of here's exactly like the compute the types of machines that we know we're going to run on in the way that some certain compute works that we can like rely on across the board if that makes sense. Other than like the most vanilla versions of these things, but I think these things are all going to I don't know continue to innovate and evolve as time goes on and so we'll be able to take more advantage of those things. >> Yeah. Uh, I'll just mention if people want uh, open source cloud for durable objects uh, Rivet is the open source equivalent. Um, there are 50 people listening that will kill me if I don't ask this question. You want CPU, they have CPU to sell, who do I sell to, who do I email? >> [laughter] >> Uh, and what do you what do you you know, and like what what is a what is a good Help them help you, right? Like what what is what is good look like? >> Well, I guess like uh, email me. And uh, >> [laughter] >> Are you sure? I don't know. [01:02:29] >> I'm going to get a lot of emails. >> But like you're you're looking like minimum fleet size, sock two of obviously enterprise like yeah, like all give me all the privacy guarantees. >> And for what it's worth like so we have a really awesome team that secures our accelerator compute um, that is now spending a lot more time with us on CPU and like where and how we can procure it. Um, to make sure that we can just continue to run our systems at the scale that we need to. And so it's not actually like my job thankfully is not to go and and run around try to um, procure compute but we do have teams that are doing that which is really great and really important. But obviously like there's a quality bar that matters to us, there's like a scale bar that matters to us. We're not going to like go and integrate and run on certain machines if like we can't get enough of it. And so all the properties that you might expect of a system at this sort of scale is what matter to us. >> Then related to the compute problem is the sandboxing problem. People have trouble containing agents in sandboxes. They keep breaking out. What's going on there? What's the state [01:03:29] of the art? >> Yeah. >> The state of the breakout recently. >> Yeah. It's interesting. One of the things that we did with Claude managed agents specifically is the way that we architected it. We put out this blog post at the time of launching where we talked about we have the brain separate from the hands. What that means is like the harness runs on a durable server really hours, right? >> Oh wow. Hey, nice nice nice one. >> Um and the hands are it's like a execution environment like a sandbox or some sort of container, right? The part of the reason why we did this was for like latency and reliability and these sorts of things, right? Like if the harness is just running on a durable server and it's only spawning a sandbox or a container to execute when it needs it, that means we can get you tokens back really quickly on your initial request. It means that your sandbox can die, but like your whole agent doesn't die because the harness is durable and over here and the you know, the agent storing state separately also. And there was a whole product development journey that we went on to get there because our first iteration of an API for agents was just like throw Claude code in a sandbox [01:04:30] and like shove an API on top. And the reason why we didn't launch that as our like durable scalable solution is everything that I just said. It was like you got to boot a sandbox and boot Claude code before you can even get a token back, right? The sandbox dies and you lose your whole agent and so reliability wasn't where we wanted it to be. So we architected this way mostly for reliability and latency. But the interesting thing there's actually a the product manager who worked on this from the very beginning was like the most important property of this architecture is security. And everyone was like, yes, yes, yes, but latency and reliability. And like now >> Now now they're vindicated. >> I literally I apologized to this person yesterday. I was like, I'm now like telling everybody who I meet about what you've been saying for months that um nobody was was as excited about as as this person was, but one of the more important properties of this architecture is you have credentials for any any work that the agent needs to execute, right? Like if you're going to go talk to an MCP server or if you're going to go like use a CLI or call some API, whatever you might want to do, [01:05:31] right? And like you need to authenticate those things. You need to store those credentials somewhere so that the agent can use them. And the way that we architected this was we have a totally separate vault where those credentials are getting stored. And as once the agent decides like I'm going to go and call out to one of these services, the agent only ever sees like a placeholder, right? And like on egress, we have a proxy that goes and like pulls the right credential out and injects it and lets that out. And so what's interesting about that is like the agent is never like in the same place where the credentials are, right? Because in the old world of like just boot the agent in a box and like let it run, you had to send credentials in there. And then you're like doing some sort of like, "How do we like scrub this and that and make sure that like the agent can't take that credential and go do something that it shouldn't be doing?" It's just like literally a like literal infrastructure wall that says the agents like can't access those credentials and can't do anything bad with them. So that was one thing that was really great. The other nice thing is like on that sandboxing [01:06:32] infrastructure, we can make sure that we have really great egress control. So we have like domain allow lists and things like that or you just like totally shut off network access so that you can't say, "Okay, agent wrote untrusted code, that code executes in sandbox, and that code can like reach out to the internet, right?" Like now it only can reach out to where you say it can reach out to. So there's a lot of really great properties around all of this within the architecture and within the way that you can configure your sandboxes that make it so that you end up like secure by default. One of the the things that's starting to happen is as we're talking to customers, we're like, "I want to run agents either for, you know, a product for my customers or internally, but I'm like, you know, hearing about incidents where agents are breaking out of sandboxes and bad things are happening. How do I think about these things? I think the two most important things are architecture so that it's like literally impossible for an agent to do something that isn't okay and the configuration you have on those sandboxes. And so one of the really, really nice things about Cloud Managed Agents is you just like get that out of the box and you get [01:07:32] something out of the box that like as we learn more and we continue to evolve the properties of those systems, you kind of get that for free cuz I think say all the words that just came out of my mouth like brain and hands and credentials injection and egress and all these sorts of things are just like if you go back like 9 months, like no one would be saying these words, right? Like these are all like relatively new problems that we're solving and so um we want our customers to be able to take advantage of the solutions to those problems as we're even learning about them and solving them ourselves within our infrastructure. >> Yeah, and internally we're just upgrading everything to be basically >> That pattern >> secure by default. Yes. Yeah. By default. >> We also unlocked something that we covered in the dark pod, which is the cloud brain local hands. >> Mhm. >> Which is projects. >> Yeah. >> Is that projects? >> Is it the rough shape of what's going on? >> That sounds right. >> Yeah, I I I feel like it's weird because obviously you're famous or Cloud Code is famous for being the super unsafe dangerous deal of permissions thing on on your machine, but now you're like finally evolving to where you probably [01:08:32] should have started in the first place. >> Basically I think like >> All these sandbox leaks, right? Are they uh a fundamental issue with our our understanding of sandboxing or container tech or are we just bad at this? Like are we are we just like not disabling like not doing the allow list, right? Like it sounds like we're just not doing the allow list. >> Yeah, so maybe a distinction that I would make is in a lot of the incidents that people are hearing about and reading about, these are incidents that are happening to models, not just all like everybody's models, right? That are in training. Where the way that you're running these models is you like put them in a box and you say like solve impossible task. I'm going to stress you out ready set go, right? And you like can only do what you can do within your environment and the agent is basically incentivized to try to get out of his box and go solve the problem. When people are running production workloads on our API or on anybody else's API, you're running an environment where like there are safeguards that are running that are like not necessarily running in training where like the whole point is to actually train the model, right? But [01:09:32] in production workloads, first of all you have those online safeguards which is really great. Second of all you have more of this architecture and infrastructure around what's happening. Like the agent isn't just in a box trying to get out, right? Like the harness is over here, the hands are over there and um it's just like not even a problem that you would have to contemplate. And so I think that some of this is just like people having the right understanding that what's happening some of these incidents that are happening in model training and otherwise are just things that like wouldn't happen in production workloads where a lot of work has been done to put the right guardrails in place so that those things like never really could happen. The second thing I would say is just like everybody's figuring all of this out like relatively newly because, you know, long-running agents is a relatively nascent thing that everybody is building and trying to deal with and so everyone's figuring out some of these things on the fly and just racing to put the right, you know, like controls in place and you know, some of that is just happening as we go. So I would say it's not you know, where people are bad at this, right? I think it's just more it's [01:10:33] a nascent field, people are figuring it out and I think when we talk about things like pacing the frontier, the way that we think about it is really specifically we're doing all the work that we're talking about right here for all of our agents internally and making sure that all these things are accessible to all of our customers so that like by default the right guardrails are just in place so that you don't end up having agents doing things that you don't want them to be doing. >> Fair enough. Um uh met well cuz you mentioned pacing, we've got to go there. We also talked about it last last time. It is superficially contradictory to advocate for pacing and then also release a new frontier model 11 days later. To me, it's just it just means that it just passed your safety checks, right? >> Yeah. Uh, and that yeah, everything we talked about earlier on like how we launched the models, um, it's like we are continuing to evolve more and more and more safety properties in place again across these like I kind of am starting to think about this in these three dimensions. It's like the actual like architecture and infrastructure security, then it's the online safeguards that are running when a model [01:11:34] is doing its work, and then it's the offline safeguards classification that's happening on data rest. So, across all those three dimensions, we're putting all the right work in place to make sure that we feel really good that we can put a model out there and it's going to be safe. And really that's fundamentally what we mean. And so, for Opus 5.5 is an excellent model. It's also really, really great on all the dimensions that we care about across those three things and therefore we feel good about releasing that model. >> So, we went for what we thought what we were hoping is the best possible user experience within this while we continue to hill climb on making those false positives better, but it's obviously like it's still a frustrating experience for people when you're trying to work with a model and then you fall back to a different model. And so, it's just something we have to >> people also understand it in the safety sense. Like you want more false positives, right? Like it's a safety issue. We clearly see the whole this is incident and you're still giving access. >> Yeah. >> To to me this is like a close cousin of the general model routing problem of do you want a smart [01:12:34] router or dumb router? Does that make sense? Like because as far as I can tell the Anthropic fallback router or detection thing is a separate model that's online. >> Uh, it's it once it hits a classifier there's just logic to where >> Exactly, right? Like so, it is not the full fable, right? You're not You're not You're not You're not running full Fable at it. And that that that's going to cause classification issues cuz it's not It's clearly not a smart as Fable, like clearly. Uh so, then there's the question of like, well, then like when I have a classification issue, that's that's like an un- justified fallback. Like, you never You never need a fallback. I'm just doing my bio homework by mentioning bio one time and then it can fall back. Uh so, like there's a question then this is very similar to the routing problem because like >> It's very similar. >> your your routing models and like how can this the the a dumb model route to a smarter model if it doesn't even recognize this That's That's a smart model problem. >> Mhm. >> Um how do we solve this? >> [laughter] >> It's a hard problem. >> I think I think in some super alignment sense, models are good at classifying [01:13:34] stuff, but not answering am I. Uh not the smartest model can tell this is a dangerous query, even if it can't make any impact on that. So, >> I think the super simplifying thing to say would be like basically you have the smarter model figure it out over to the one that um it needs to get to. >> I think you also >> It's slower at cost. There's There's trade-offs there. >> There's trade-offs. I'm not saying it's like the right user experience, but it would probably be a slightly more effective route. >> thing interesting is you also don't cut off immediately. I've seen cases where it will start answering, start reasoning, and then get hit by a safety filter. So, my first example with Fable was throwing the Fable safety card. >> Mhm. >> And it started re- I was like, just break down anything interesting in here. Starts breaking it down. 5 minutes in, I think it gets into cybersecurity and it's like >> Yeah. >> Oh, we're cutting or rolling back to Opus, but it did a lot. There's a lot of context, and I see it. I It got to a point where it's all about cybersecurity safety, and then it rolled back. And it's like, it wasn't rolling back my query. It was the combination of [01:14:35] reasoning and the whole chain, right? >> Yeah. >> Do you have a idea of what the >> Yeah, let's brainstorm what what what's better. >> I think the user experience and how we resolve that user experience is probably, you know, a separate question. Like, maybe it's a new a super smart model to handle it or or whatever. Um, but I think that the more appropriate UX that I would imagine for a user is to actually have a more consistent unified experience with Claude, where instead of getting you falling back and then you can double fall back. There are like logical points um in our tree that that do do that for again, like say a few reasons and whatever, but I think the UX is just like really convoluted for a user. It's like you almost you you're trying to do something and I I would probably take the case of like good user. And you're a good user, you're just trying to do something like cyber bio homework, like assess the cybersecurity like safety card, totally reasonable. And you get blocked and you fall back. And I think instead it would just be like you are where our model should have gotten or classifiers or however you want to do this would be smart enough to detect that that is a situation that happens and just write down the the answer for you more concretely while [01:15:36] dropping down the nature of the like uh negative use of that particular capability. So, going back to that bio homework scenario, while we may have classifiers that have detected that it is possibly bio. >> I'll also say like the on the API, we went with a spectrum of solutions for this for customers. So, first of all, like if you've built something on the API and you run into a block, like you don't have to like continue the conversation on a fallback. That's what we've done in our apps and we've made that a feature of the API and you can do it like server-side within our API. You can choose to like set I want to do the fallbacks, right? And like we'll do that. We also have it within our SDKs on the client side. There's helpers for if you want to like manage fallbacks in various ways or you could choose to like re-prompt differently in some way, right? And and continue with the same model. Um, but I do think that there's some in between interesting solutions that we'll continue to experiment with to figure out how we unlock people and and do so safely but with a good user [01:16:36] experience. >> Yeah. >> And then you brought up the the beta that was like very similar to routers. I didn't go like I totally agree with that. I think it has a similar UX problem as well. Like with a router you you go through it's like let's say you want to do a cost router right for the task complexity that you have. Makes sense. Very hard to do. You don't know everything up front about your task so you can't like even with perfect you just don't actually have perfect information. Even if you could actually understand each model perfectly which you don't but even if you did you also don't have perfect information about your task. And so there's a lot of error associated with your your routing logic that you would have to create whether no matter how good a classifier you produce and yeah maybe it's like a super big brain model type of problem but regardless it's I think a very difficult one. And in each one of those like scenarios if to find a way to like route that properly it's very very convoluted. You could expose all that logic to customers like hey you know your task was to center a div so I'm going to route you to something super cheap and like then this other one's like detect vulnerabilities so I'm going to route you to something like really expensive. If you expose all of that I don't think that's like a great user experience ultimately either and for that similar [01:17:37] class of problem I think the more idealized UX would be something along the lines of like dynamic like effort or dynamic like understanding of the level of complexity of your task which you might actually evolve over time as you learn more and more information about the task. But in that world if you're kind of like moving between a lot of different models it actually becomes very very disjointed and so like maybe a more unified approach would be something like it's a consistent model but on it's like very very low effort level or something like that it's actually very very cheap and very very affordable but it can scale itself up very very continuously but you have like a very very consistent persona towards that model. And obviously speaking about this in the context of like you know like our family of of models if you try to even do that across other family of models other model providers like the personality the disjointedness like starts becoming very very complicated. I don't know how you solve it in that layer but at least within like our family of stuff like I'm more bullish on things that are just more consistent and there might be ways to solve it more robustly for customers. >> Yeah you're you're not responsible for anyone else but within Claude yes yeah I mean this is your your territory. [01:18:38] >> Yeah. >> [laughter] >> It's nice to see the way you guys think of these problems, right? Intent in and of itself is a hard problem. Like, I feel like the models have gotten better, but sometimes people don't do a good job at classifying what they want, or they just slide her all the way to one extreme, or, you know, the model doesn't really understand. You're asking a simple question, and it's like, "Oh, I have a whole HTML landing page now. I just expected a few line output, you know?" >> [laughter] >> So, there's always a a measure on there. >> Yeah. I think we we in in this conversation we really talked about a lot of safety. We talked about like watching the models, like all this kind of stuff, but it's obviously in in pursuit of helping power the workloads for people. And, like we discussed the kind of the earlier part of this conversation, we are seeing workloads uh change quite a bit. I think like in some workloads intelligence does seem to be increasingly saturated, which is like a a new uh situation. Um and it's it's getting to the place where I think there are some class of tasks that are still very very unsaturated. I actually think they're more than maybe people anticipate, but they're increasingly going into the direction of you get that [01:19:39] alpha um or that return by doing extremely long-running agentic tasks, and things that historically you were not really able to brute force, but maybe you can with a lot more intelligence behind it. Um and those workloads are really really fascinating to us, and we see a lot of value in that kind of area, both from businesses getting that value from themselves, as well as, you know, like new companies or startups, or however you want to frame it, um who are interested in building kind of like agentic products on top of it. I think if you look at the class of things today, a lot of it tends to be So, obviously a lot of coding stuff. There's obviously a lot of like enterprise uh innovation acceleration kind of stuff. Um but they tend to take the shape of automating a very distinct and specific process. And I think in some areas that's actually really starting to work, which is awesome. In some areas it still isn't, so there's still a lot of like room to get to kind of evolve and push the frontier over those models. And then there's like this new class of workload, like the the bio stuff that we we about, or some of the cyber stuff. Like, that's a it's just a very, very complicated set of like I would say hard engineering or hard science problem. Um and those workloads are increasing when you're sharing up as well for us as [01:20:40] like a really interesting space to just deliver value to customers. >> You know, you you are at one of the few spots in the world where you're discovering all these workloads and we're automating it. It's like >> Wait, that's pretty crazy. >> [laughter] >> Uh I mean, you know, that's um very inspiring. Uh you guys are having so much impact and and I'm really uh grateful for that. Uh we're going to go into a few closing questions which I which I thought would be fun to sort of bounce back and forth between the two of you. Um how do you fight slop on your teams? >> Uh So, we actually this has been an interesting debate within our team. The the we're no different than anybody else, right? I think plenty of teams are talking about this. Um and what's interesting is what we're trying to figure out is if you're producing a document, you're producing something that you want other people to consume, code, documents, whatever, right? If like you should use Claude to help you, of course, right? But if you use Claude to like totally generate a thing, you didn't spend any time on like actually reading it yourself and making sure that it was something of high [01:21:40] quality, then you're putting a whole extra burden on the consumer of that document or of that code or whatever to spend a whole bunch of time trying to parse it and that's just like insane and totally unfair. But what's funny is we you know, we push back we push on our team like you can't do that, right? Like if you're producing something, use Claude, sure, whatever, but like you have to make sure that it is a high quality thing. >> Yeah, but you as the human are still responsible for the quality. >> Right, you're [clears throat] still responsible for the quality, but then you get into this really funny thing where because of the Claudisms, even just like a silly like I don't know, there's like an M-dash in the middle of a thing, right? Like people's brains trigger on this like, oh no, I'm reading something that was written by Claude and not a human and I'm like annoyed because I want to be like reading a thing that was like fully produced by it. No, it's really fully produced by a human, but like that has the quality bar that the human would hold, right? And so then we get into funny situations where maybe actually like the person did spend a lot of time making sure the thing was high quality, but they used Claude to help them, but there's those little hints of [01:22:40] Claude and then the reader is like turned off, you know? And like that's like unproductive as well. And so what we have kind of tried to do is like set a bar within our team that's very much like use Claude to help you because it would be really silly if you didn't use this hilariously powerful agent to speed you up in your job and in your life or to just like help you produce better things, but it has to like have a certain bar. And then for the consumer and the reader, if like it really actually is just slop, like send it back, right? Like just say no, just send it back, like you're not going to waste your time on this. But also don't be like too nitpicky about the little Claudisms, right? Because like we have to find the right balance here. And so it's just really funny because we're, you know, we're going back and forth on this stuff within our teams. And just like so is everybody else in the world right now. Um we're just not any different. >> But but you have free tokens. >> Uh well, sure, [laughter] but um I mean it that doesn't tend to matter quite as much when it comes to like >> As as as in so you have more than anyone else in the world. >> see a lot you guys use a lot. You see a [01:23:41] lot more people doing a lot more of >> [laughter] >> Yeah, it's totally fair, but it's definitely it's definitely >> heavily discouraged, yeah. >> Uh engineering stuff? >> Yeah, I mean I same thing, right? Like but but I do think that where where we've gotten to is a pretty good spot where like you know, humans play a large role in making sure like designs of systems are good, right? And like implementation tends to be Claude, right? And like we're in a pretty good spot where for like a lot of the code that gets written, like all you really need is just like a quick stamp and you can keep going and like you don't necessarily have someone like deeply reviewing every single line of the code cuz you care more about like the overall plan, the overall like design is in good shape less so than like the exact perfect implementation, and obviously rely on a whole bunch of verification to make sure the thing that you're going to ship is high quality. And so, it's actually been like less of a conversation these days on like code quality being sloppy in part because Claude just like gotten better, but in part because we have a lot of the right verification in place to make sure that [01:24:41] like, you know, we're holding a high bar. >> see. I see. Trust your tests. >> Yeah. >> Um any product PMing slop, you can ship features a lot, you can put stuff out in the day. >> he said this to me recently. I did Do you remember what he said? >> All right, all right, tweeted this. The right way to use model capabilities is not to ship 10x more features into prod. Uh spend more time understanding your users, trying experiments, building, whatnot. But basically, you know, you could very easily just keep shipping random slop, but you you should actually think about it. You should spend more time kicking off that first prompt. I know now today he's like, "Kick off plan mode. We don't need it anymore." But uh outside of plan mode, you should think about what you're doing, right? It's very easy to put out a lot, and you can get product slop. >> Yeah, I mean, I I tend to kind of go on this rant about There's a lot of just like fundamentals about how software gets like and products get planned and shipped and otherwise that like doesn't need to change just because agents are like helping speed up certain parts of the process, but there's this inherent [01:25:41] like almost laziness that's starting to come out from a lot of people where it's like, "Oh, I don't have to like think about some part of the process at all because the agents are doing it and helping me or whatever." But like, those are still like incredibly foundational parts of the process that like should be things that you're spending your time and energy on, and I think that's kind of the spirit of um what he's getting at here. >> It's just like as a founder or creator, let's call it as well, um it just feels like because the cost of experimentation or exploration is so low now that I should be doing more of that. And he's basically saying to not do that, to to like use it in service of a a very intentional goal that I start from the start. Which like I obviously I'm coming around to. It's just hard. It's like you have this toy that does everything. You're just like throw it at the wall. >> You know, there is scope, too. So, like the blog post of today was 2 weeks to make Cloud AI faster. There were 12 things. Uh we had There were 13 things. We hit 12 of the 13 targets in 3 days, right? [01:26:41] Well, you could just stay, "Okay, it's a 2-week sprint. We'll do the 13." But instead, there's 20 things you push bound. This is not product slop. It's not new products. It's refactoring. But still, you guys do a very good job at outputting a lot but keeping high-bar quality. >> Yeah. Nice. >> Okay, part two of three. Uh this is going to make Angela very comfortable. This is Angela tweet corner. Uh I will tell >> [laughter] >> Nice. >> I don't know. >> I will I will I will tell people most of to Uh I I and I I I'm pretty sure you've deleted stuff. >> [laughter] >> Because I'm like there's there's stuff there's stuff in your deletions that I'm like >> That you deleted. >> Yeah, I know. It like really like altered my DNA. >> Oh, I appreciate that. >> What do you refer What do you think about the most? What do you refer people to most that you want to sort of bring up? And I can I can pull it up for you if you want, but >> Um I actually I don't think I have a favorite, honestly. >> Oh, come on. >> No, I really do Okay, I guess maybe I like the one that you say where it's half doing something is an expensive way of not doing it. >> Yeah, that's that's that's very on topic for us today. >> Okay, nice. [01:27:41] >> [laughter] >> But I guess you know, you obviously you're in a different job now and you don't do this anymore. But like that that I think this is a power tweet. This is like a Naval Ravikant. This Naval is you. >> Yeah. >> That do this. >> Uh-huh. >> And like what mode What mode of what like what framework of mind frame of mind are you in? >> That's a really good question. Yeah, I know. This This was a very long time ago. Um I went through a period of my life where I was like reading constantly actually, I found myself like very bored um of the stuff that I was doing and I felt like I was just like not being like intellectually challenged. And so I actually went on this a bit of a bender where I I think I read something like 200 or so books in like a year um and I was just curious about like every I just wanted to like read everything and I was missing a little bit of like the the kind of like intellectual like I don't know like muscle building you get when you're like in school. Where they like do you like this is a new topic you [01:28:42] got to learn it's like tough. It's like not super fun actually but you are really like stretching yourself. And I felt like after just like working for a couple years I just like stopped doing that. Like yeah the job's like interesting and yes it's like challenging but doesn't have like that like intellectual bit to it. So anyway went on this bit of a a bender uh a reading bender uh and I like read like 200 bucks in a year it was great. I had like a really long commute so it was like it was a good use of my time. Very long like Caltrain ride but uh I ended up coming out of that and being like I was just like thinking a lot um as a result of that and at the time it really wasn't a build like a Twitter follower or anything. I was just like tweeting into the like abyss. There was like I have like no followers. These were like tweets for myself and I was like practicing a little bit of like how would I try to compress the idea that I have in as few words as possible. It was kind of like the personal test I had for myself. And then I guess people just started liking it and so I just I just kind of just kept going at it. And then I got too busy with work and I I stopped doing it. So >> I got to ask after all that reading any book book recommendations blogs long [01:29:43] articles? >> My favorite one still is probably The Black Swan by Nassim Taleb. Um that was like one of those like I was like to me at least it was like a bit of a like a life changing philosophical book where I was like that's a completely different way to look at the world and >> You know speaking of Black Swan when the favorite thing happened right? It's a thing that you didn't know it happen. Now that happens like it's a probably changed your framework of >> Exactly. And I guess like so simple, but I think you did a great job of just really hitting it home and articulating it and how it's like a common property of actually of so much of the world. So I really love that book and that's probably my favorite. >> Uh and for listeners, if you're bored with that one, you do Fooled by Randomness Randomness also by Taleb. >> Did you like Antifragile? >> Not so, not so. The the more famous he got, the more intolerable he became. [laughter] But also question for Kaylin, my if you do have one, do you have a similar like this is a book I recommend to everybody? >> No. That's a good question. I need to [01:30:43] think about it. It is funny cuz I do think we have very different styles on this front. Like she's no longer like in fortune cookie tweeting mode, but I do think that like you stopped that, but I do think a a big part of how you work and how you work with other people is like philosopher mode sometimes, you know? >> fortune cookies in Slack? >> Oh, I I do actually. Not intentionally. I think it's might as well be. So when I >> But I think that's a core success of succeeding in topic is that you have to be very good at Slack. >> Yeah. Oh, you have to be very good at Slack. It is it's a pretty intense place, but I think I'm a little So I'm kind of psycho about like inbox zero on both email and Slack. It's easier on email these days than Slack. But I like pretty aggressively prune. Like here are the channels that I'm like actually in and paying attention to versus the ones that are like muted below the line sort of thing and I'm like always making sure that I'm on top of the ones that I need to be on top of, but it is it's basically it feels like a full-time job. Like there are days where I'm like I could cancel all my meetings and write no documents, do no work, right? And literally just sit there like [01:31:44] on Slack the entire day and I would have done what felt like an entire day's worth of work just by like engaging on Slack with all the things. It's pretty wild. >> No, but that's horrible. Like what what do you do? >> how I feel. I feel like you do all this you put in all this effort and yes, like you met the mark on like responding to people and getting certain things over the line, but you didn't actually produce the output that you wanted to. And like that's like a that's like a personally very hard thing for me to deal with. So, I try to be like, I need to like dedicate time to like actually thinking and actually like trying to produce something in order to feel like I'm like actually >> And ignores and drops stuff, right? You you're going to drop stuff. >> Um yes, I have gotten a lot of feedback I'm a little hard to reach on Slack. Um yeah, pretty terrible >> What ends up happening now is I get a lot of Slacks saying, "Can you get Angela to respond to >> [laughter] >> And I'm like I'm like really trying to write back to people. >> Wait, wait, so do you guys don't have like a personal Claude like assistant something to There's got to be a mom, but there's Why has no one no one written in personal EA for each of you? >> Yeah, well, so we do have we have a human one also. We also have We have [01:32:45] plenty of Claude help in Slack for sure. Um but I do think actually the the types of Slacks that I'm like when I'm printing through and getting are often just like our team is like blocked on something cuz there's some hard decision that needs to happen or whatever it is, right? And it's like it is just like a form of whether it's in Slack or some other way, right? Like it's a form of getting the team unblocked like executing in a high quality that is my job anyway and it like happens to happen a lot in Slack, but I do agree that it ends up feeling like, "What did I even produce?" when you get through a whole day worth of literally just like going back and forth with people and trying to solve problems. >> You you produced the a functioning team. You >> [laughter] >> that's unblocked. So, I will give you a props to that as a as a former manager that is Yeah, yeah, yeah, you don't feel like you produce anything, but like you produced the team. >> Yeah. >> Uh speaking of the team, I was going to go to you next Caitlyn. DevRel Corner. >> Oh, nice. >> Uh we are I know you're really hiring DevRel so I'm going to make a call out for for developer relations. You hired Addy Osmani recently. [01:33:45] >> Yeah. >> Uh great get. Lance You have you have you have sort of been on the team. What what do you guys want to hire for? You know, this is this is it to talk about it. And I'm happy to trade notes. >> Yeah, >> well, props to you because I will say when >> outsourced Anthropic recruiting team. Recruiting team. >> Literally. >> [laughter] >> Um >> It's fun. >> And it's It's actually not even just recruiting. Like I'm going to toot your horn a little bit. When So, I when I joined Anthropic, we, you know, we're a small team. We're like, all things considered, we're an extremely small team. And so, we run the platforms, the API business, and and all of our product infrastructure. And we we had like a very nascent DevRel function within Anthropic that was, you know, like not necessarily within the product and engineering org. And when I joined, we were like, "Okay, we like need to grow up a little bit and need to figure out how to like engage with the community, engage with the world." And um we need to like really invest in this DevRel function. And we ended up spinning up a DevRel team within platform and a DevRel team within [01:34:46] Claude Code. In part because it's like two pretty different like developer ecosystems that we're trying to engage with across the API business and Agent Coding, right? And so, we ended up doing it as like two separate teams, but kind of like dotted line to each other and spend a ton of time together. Um but it was it was something that I was going to be responsible for like DevRel, which I've spent plenty of time with DevRel before at Stripe or otherwise, but it was never something that I was responsible for. I was like, "I've no idea what I'm what I'm doing." And you were like, "I got you. Like, here are the three pillars of DevRel. Like, here's how you think about it." Which is really great across like product, content, community. And then now I can like rattle that off to people and they're like, "Wow, you really understand DevRel." I'm like, "No, Tom taught me." >> [laughter] >> You're like, >> Yeah, Tom taught me that. That that was not a good >> It's apparently that post is apparently like the default reading for Google when when they onboard new DevRel. It's like really funny. >> Yeah, I saw. Um it's it really like this this really helped me because I had no idea what I was doing. Um but then on top of that, we, you know, like I've I don't have a manager for the DevRel team within platform [01:35:47] right now. And so, that's one of the things that um I'm hoping to hire in our DevRel ICs, um, report to me directly, in part because I think it's a really, really, really hard problem right now, but also in part because I find it really fun. And I'm having a lot of fun working with the team. But really specifically, like, what's hard on platform is platform in some sometimes in our like public narrative, like, our platform business is a really large business, but it can kind of get lost in between research and policy and all this work that we're obviously doing that's like our most frontier important work. And then our applications, like, Claude Code and co-work and all the things happening over here. When you think about Anthropic and you think about our public narrative, those are some of the things that often come to mind and less so that we're running this like massive developer platform business that's really important and so many people are building on top of. And so a big goal that we have with our DevRel team within platform is figuring out not only like the really classic like, how do we better engage with, like, you know, product, like, what people are building [01:36:47] and how we get feedback back into the product and these sorts of things. of it is just like finding our voice and finding what our brand needs to be and what our narrative needs to be and like, who we even are going to be in the community and not kind of get lost in all of the things that are happening. And so, that I think has been interesting cuz I think the brand that we've started to spin up for our DevRel team, in part because actually Lance joining our team was you connecting me with Lance, but Lance is highly technical in the way that he goes about DevRel is very much like teaching, right? Like, he writes a lot about memory and managing costs and these sorts of things. And he goes really deep, he builds evals, he like has a lot of really technical content that he puts together. And I really like that idea, I like that idea of our our DevRel team being kind of like thought leaders for people within what is a really nascent, really hard, really evolving space. >> Educators, thought leaders, explainers of things. Um, there's another version of this I really like called the master builder, uh, which is I will build the things with the platform going you what's possible and then you'll be like, "Oh, I didn't [01:37:47] know that was possible with with Cloud Platform." And then you you just go. And this is a thing stolen from LEGO. Yeah, because LEGO, how do they sell LEGOs? They have master builders who build things with LEGOs and you're like, "I want to do that. I want to be like this person." And then they they they do that. >> Nice. That's awesome. Yeah, and the CJ, the other person on our team, who's really great as well, I think he also does a lot more of that sort of like >> Yeah. >> um you know, writing cookbooks and like putting out examples and really engaging with the community on it. >> Cookbooks is funny, right? Like when APIs launched in 2023, as called it for LLMs, very important. Now >> Not so much. >> [laughter] >> Yeah, well, it's interesting cuz you're less in the spot where you'd like copy-paste the code and like use it, right? Um but I do think it's like it it shows the art of the possible for people of just like what even is the use case, right? And like how could you go about accomplishing it? But I hear you that like, you know, the code itself is maybe less Like we for the longest time had this like there's there's an engineer on our team who wrote our [01:38:47] cookbook on computer use. And it like haunts him to this day because it was like massively complex and like, you know, kind of a beast of a thing that I think, you know, a lot of people got use out of at a time when it's like, "I'm going to build a computer use implementation and I'm going to deal with all of the the screenshotting and all the things, right, that are happening between the messages API." >> Yeah, we were there for the hackathon that you guys launched in our office. >> Yeah, yeah. And like though that was the era when you like needed to have that sort of thing and now, you know, it's much more about we have a skill like the Claude API skill. It's like really, really excellent at helping you either within Claude code or any other coding agents to build a really high-quality integration. And now like these sorts of things like I'm going to build a computer use integration, like how do I go about doing that? Like very few people I I think {slash} hope are reaching for the big gnarly cookbook and more so they're loading up the skill in a coding agent and working that >> Yeah, or they're just putting it into a reference architecture. Yeah, it actually does much better. Like increasingly we're seeing that for a lot of the things that you might want to have different implementations on uh [01:39:49] typically at the harness layer, but there could be other variants. You just point Claude at and you say like this is the reference architecture I want. >> Yeah. >> And then you build it. >> Fun fact, the Claude API skill that helps people build against the API is owned by our DevRel team. >> Um so that was built by Lance, right? Exactly. Like that's the sort of thing. >> high-leverage form of DevRel is to like distill everything that is a best practices that's going to go and go into those agents. >> Yeah. >> One of the reasons it's weird because like I'm regarded as this like DevRel person or authority, but I like it's this like there's some part of me that just wants to tell you you don't need DevRel. Get so good you don't need it. Right, or right, you turn the engineers, the the the leaders into the public faces. Because all you really you are trying to do is public faces and like here's how we do things here and then that's kind of it. You something something like that. You're you don't need like, you know, to people hear about you. So like there I think to me when Boris came burst onto the scene like it seemed like a very like hey, I'm Boris and the creator of Claude code, blah blah blah. This is how we do things, blah blah. And you like yes, he [01:40:50] has the day job, but he's also doing that functionally. So like why why doesn't everyone just do that? >> Yeah, I mean it's a really good question. >> [laughter] >> And there are teams that have done this really well. Like I think the Cloudflare team is a really great example of like >> yeah, all >> They're all awesome, right? >> Nonstop. >> Yeah. >> And to me that's actually a surprising a very visible sign of a good manager is that they all have their jobs they're all like publicly identified. They're not quitting. >> Yeah, totally. It's interesting because I think this is actually something we are trying to do more of with the platform team and with our team and and folks who, you know, can kind of just get out there and like be part of the community and and be speaking and be on X and be wherever they are, right? I think the problem for us is just like having time and kind of knowing the right way to do that stuff at high quality. So I think I see the DevRel team in some ways as just being like the experts in doing this sort of stuff at quality. They can do it all day, every day, and they can help enable the rest of the team to think about how to do this with the small amount of time that they are able to spend on this sort of [01:41:50] stuff to do it at high quality, too. >> Yeah. So, like yeah, so I call this like kind of the unbundling of DevRel into a front office and a back office. So, like a typical person who does this job does the whole thing, start to finish, they're kind of autonomous. But, actually [clears throat] you can sort of build out this back office that lets your product leads and their engineering leads actually take the spotlight with a simple contract of like, "Okay, you sit down with me for 30 to 60 minutes, and we'll do the brain dump, and then you're done. You go back to work." >> Yeah. >> And then we'll like we'll like ghost ghostwrite for you, whatever. But, like that team of like production people, writing people, whatever, is sort of provides a service to the rest of the rest of people. And I I think that to me that's like kind of the future of of this of this stuff, to be honest, because people like don't actually really want to hear from DevRel. They want to hear from us. And and like, you know, >> Yeah. >> And you have like lots of Tariq you have Tariq. >> Yeah. >> You know, and Lydia You got Lydia through bun. >> Yeah. >> Exactly. But, like he he like functionally he does the job. He does the job. >> Mhm. [01:42:50] >> And so, like that you want more of those cuz people people want to hear from people in the products, not not like who are like Pete smoke Pete shows I Pete Pete spokesperson. >> Yeah. >> [laughter] >> Which I get a lot of, and that's fine. That's a job. Somebody has to do it. And so, then like reasonably you can expect people who are working on engineering to not be good at it. >> Yeah. >> So, like you should pay someone who specializes in it. Like specialty is is a skill. Like if I think there is a back office front office separation that that can emerge. >> Yeah. That is really cool. >> Yeah. And my analogy would be like everyone hires, but you have an HR department. HR department enables you to hire. You hire. Like yeah. >> Makes a ton of sense. >> [laughter] >> Cool. That was my soapbox. You know, closing questions. You've been very generous with your time. Any parting thoughts or calls to action? >> You should build something cool on Opus 5.5. >> Yeah. People should try to push the bounds with new Opus 5.5. You guys have been using it for the longest. >> Yeah, I mean, it's really cost-efficient. Um, it's less verbose, all the things we mentioned. The vibes are really, really [01:43:51] high. It's been fun. I yesterday was a really fun day for us cuz like finally again, I got like a whole slew of texts from friends and and people who were just like, "Amazing model. Thank you." Like one of my old bosses messaged me and he was like, "Opus 5.5 just saved our butts in an incident, right?" And it's like, "Okay, great. Like we're we're back." is really the energy that we feel. And so, I think things that like maybe you wanted to experiment with something and like cost was something that was front of mind for you, right? And like token efficiency and these sorts of things like that kind of might have held you back from like really like firing it off and like letting a model run. I think that's the Those are the sorts of problems you should return to now with Opus 5.5 cuz I think you'll be able to get just like a lot further than you maybe otherwise would have been able to with models that were less cost-effective. >> I think you should try something with like a multi-agent setup. I think you will find it to be very good. >> Any any other hints, by the way? I mean, multi-agent is very broad. >> Uh, nope. I think it'll be very curious to hear what people build. Um, but yes, that it's uh it should be extra good. [01:44:51] So. >> Yeah, including, let's say, roles that are not strictly defined, but they assign among themselves. Right? Which is uh to me the that's the new thing. Like you can assign roles, you could always assign roles. You can give tokens jobs. But uh this is like they're the giving themselves jobs. >> Yes. >> I guess um say more on multi-agent, right? Like should we use sub-agents? What do you How should people think about multi-agents? >> This is pretty good in the sense of like basically generally you want to have one lead and you want to like split up like a bunch of different stuff. And as long as you can have a task that you think could be generally parallelizable, and they share uniform context across it, I think you will get a lot of gains. Now, of course, it depends on the problem that you're pushing at, but like increasingly this is a very superior way for us to handle a lot of these things, and especially if every sub-agent is, or whatever you want to call these things, sub-leads, they are asynchronous, and they're not blocking of each other, that is like the only two bits I would say you should care about in your [01:45:51] architecture. Other than that, like let it rip, and like see what Claude designs for you. And increasingly I think a lot of this architecture stuff will fall into this direction. Like Claude will get smarter and smarter. It's going to learn how to figure ways to like split itself, quote itself, adjust itself to the architectures that make sense. You just need to prevent it. You just need to make sure it can like parallel process. So, it needs to share state, and it needs to like not block its sub-agents. And as long as you contain those two principles, honestly, I think people should experiment, and probably try some really crazy stuff. And I wouldn't be surprised to see some like really interesting things on top of that, that maybe they felt like there were some intractable problem that was really annoying and that keep pushing for, that something like this could actually probably just like crush and maybe one shot or something. >> Excellent. All right, everyone go build. >> Cool. Cool. >> [music]