The enterprise AI stack behind Stripe’s company brain “Kai” transcript

Agents are very creative at bringing your infra down. It turns out that agents just like dial up all your failure mode. It just multiplies the amplitude of problems you can get. There were agents that went rogue. There were agents that may have almost taken down core systems. One of the cool things about projects that can be very concrete for people is the idea of tool policies. Let's say you're a person on the HR team who's dealing with a bunch of sensitive information.

You really don't want the agent to sort of go rogue and put that sensitive data into some public Google document that all Stripes can access, but you also don't want to tell them, oh, you can't use any tools because your workloads are too sensitive. >> I love this idea of this like three layer triage that a data agent can go through and that's really smart. Find existing reports, then use the analytics layer to find the right query.

Then if like you really have to fall down to the data catalog and write your own query. Your data warehouse has to be very resilient to high volume queries [music] because when in doubt an agent will just brute force it. >> We could test this out and it's going to kick off what is a human in the loop workflow. Create a calendar invite for me and Wong tomorrow at 11:00 a.m. Pacific. This isn't the fun part. The fun part is what I showed before.

Agents are really good. They're very creative. So, we got to put some restrictions on them so they don't go rogue. [music] >> Welcome back to How I AI. I'm Claire, product leader and AI obsessive here on a mission to help you build better with these new tools. Today we have Sherrod, an engineering manager at Stripe and part of the team who built Kai, their internal company brain and company agent. He's going to show us why you might want to build your own custom agent for your company.

What are the governance and control mechanisms of Kai that make it super special and how to build not just a skilluing skill, but a skill building platform for your team to share their automations and workflows with the rest of the company. Let's get to it. This episode is brought to you by DX. [music] In a recent study across more than 500 engineering organizations, DX [music] found that spend on AI tools has grown 28x over the last year.

[music] The share of AI authored code is climbing, but overall innovation has remained flat. As teams generate code faster, new friction in code review and validation is offsetting those early [music] velocity gains. DX tracks speed, quality, and cost together across the software development life cycle, giving engineering leaders clear visibility into how AI impacts delivery and whether those investments are translating into real value.

Download the full report at getdx.com/howi [music] ai. That's gdx.com/h how I AI. Shared, it's so nice to have you here. And I [snorts] had to reach out to the Stripe team because I wanted to learn not just about how Kai, the company brain, the company agent works at this big company, but I really wanted to understand why in the world you built it yourself. And so I wonder if you could just start us there. Why build Kai?

What was the problem you all were trying to solve? We wondered about this a lot before we built it because we had all this like the this avalanche of AI tools. This was like early 2026. Huge number of tools coming out. Plot code had just like taken the world by by storm. Like a lot of like really cool things were happening. And the problem that me and my um and my colleague Anupam were dealing with is how do we get AI to everyone, right?

And we quickly realized it's not just an engineering or a technical problem. The harder problems are in trying to replicate the way a company works at scale and stripe is an incredibly complex business like all around the world like multitude of products so many so many processes that keep uh keep us in a shape so that we can help our users and so we quickly realized that it's not about providing AI it's about providing the correct governance structures so that everyone can just go use AI and know it'll do the right thing for them.

So some of the things that uh we really thought about and why we built Kai is uh governance is a is a big thing for us like I just said. The second thing is Kai is like really interesting. It's context aware. It knows who you are and what you do at the company and it knows what are the things that your colleagues are uh mostly interested in and it has access to the orc chart and your projects and all the things that are going on which means that it has a lot more uh mechanisms to do the right thing for you and understand what you're trying to do.

Um and the other thing that that I really like and our security team really likes is that Kai is all hosted on the cloud. It's always on. It's behind our standard security boundaries and uh it's very stripey in that everything that we use to build Kai actually ends up being standard infrastructure that helps us build great agents for our users. And ultimately uh while it's great for me that like we're really helping Stripe be more effective and enabling everyone.

I really am happy that what we're doing here is helping build the rails that make our stripe users uh get better products uh out of this using you know great agents that are coming out. >> So just to kind of repeat back what I heard and some of the unique things that you built into Kai is one this sort of bounded context engine. So you know I'm Claire, I work at Stripe. Kai knows about me. It knows about my place in the org chart.

It knows does it know kind of like strategic projects that I'm working on? Does it know like conversations I'm having? Like how how do you how do you ingest that knowledge? What does it know about an individual? >> We allow people to select how much they they they give Kai access to. But out of the box they Kai knows like who you are and where do you sit in the art chart, right? And um it also knows some helpful other things like like what day is it and what time is it and so on. relevant to you.

Personalized context is like who you are by use of the orchart. From there on, the standard tools that we have connected to Kai let you talk to our uh um our project management system that figures out like all the OKRs and all the recent shipped emails and all the projects you're part of, right? And it lets you connect, if you choose to let it, to your Google Drive, to uh to your Slack, right, to private messages, things that are like pretty sensitive uh and uh we really like keeping a keeping a a tight boundary around.

You can choose to let KI know as much as that as you want to. Um, some people choose not to, and I'm I'm actually one of those. I I I I turn off and turn on and turn off my access every uh every session or every day. Uh, but many people are like, let the AI figure it out and more context is better. >> Great. So, you have this context engine but controlled by the end user to to some extent. So, you can decide as an employee how much you want to automatically ingest into the system. um you have these you're also building it on some fundamental agent building blocks and you know this is one of the you know there are lots of reasons you might build one of these things um at a company one is just to teach your team how to build great agents.

It's like the perfect dog fooding agent experience, which is if you're going to build a great agent for your customers, you should learn to build a great agent for yourself. And then it sounds like you're reusing some of that infrastructure, which is nice because then you can pressure test it against Stripe employees or or external customers. What else is unique about Kai? What are the places where you're like this we we went a little extra on building this so as folks that are listening thinking about building this internally they can decide where they want to differentiate their own kind of internal agents.

The two things that I'm uh we were very intentional about is the idea of projects and projects are primarily a governance mechanism but they also let you you have this context engine as you said right but projects are almost like intentionally the user is telling you what they are trying to do and that's a very strong signal of intent and that lets the uh AI perform a lot better but projects uh you can if you if you see my screen in a second um there's the how I AI demo project that I have right here projects are some things that stripes can create.

Um it's it's all open, right? And what they can do is uh we have projects that have 500 people in them. We have projects that have five people in them. The idea is that you create something where someone decides what's the appropriate set of things that you should need for the AI to function well and what are the appropriate safety controls. Like token spend is top of mind for a lot of companies. A project can kind of say like hey here's the default model we want people to use. we don't even want to let them use these like super expensive models, right?

Because the the job that you're trying to do here doesn't need, you know, uh one of these super models to uh to look at them. So projects as a governance mechanism and really um I think we did a lot there because again, Stripe is a very complex business, enterprise scale AI requires these sort of mechanisms to make sense of things. uh and I like the fact that we can have a few people who are uh very you know knowledgeable knowledgeable about the EI and the trade-offs between cost performance and uh latency and they can sort of set the stage for everyone to just go use right we should try to minimize the number of people who have to actively make these choices every day and just it should do the right thing for them and projects are one way that uh we can do that um the other one which is kind of related we'll see that as well our skills the way that we build skills and or skill routing.

Uh there's a lot there and how we think about skill quality, skill governance, a lot of good stuff there that uh that we can get into uh that we've invested a lot in. Uh the thing I'm I think the team has shipped. It's going to be interesting. It's going to be a live demo, but I think they shipped a feature where automatically people uh everyone who authors a skill gets suggestions on how to make their skill better, how to help client uh built into the platform.

I I like this because I've seen, you know, we've seen projects for holding context or just simply as an organization structure to AI work, right? Like here are the files you need. Just put all my chats in this group. What I haven't seen anybody talk about, which I actually think is really interesting, is using projects as a configuration layer on and a governance layer on how your team actually uses AI to get a specific job done.

And so I like that idea of project-based model routing. project based I'm sure connectors or approvals or you know all those sorts of things and I'm sure there's more and more and more you could do in the future. Um it's really interesting. So let's I mean show us what's Kai good at and why don't you just walk us through a a common Kai task and how it how it how this customization benefits the end user in particular maybe like someone who's a little less technical. >> Yeah.

Uh absolutely happy to do it. So uh I have a few trusty prompts here. What we're going to be doing today is we're actually going to be creating a dashboard because Stripes, we love our data. And something that literally everyone at Stripe uh does with Kai is create a bunch of dashboards, right? So, we're going to do this live and we're going to ask Kai to create a dashboard for us. And uh you know, in the interest of vanity metrics or things that I can speak to uh easily, I'm going to ask it to go create um something related to Kite itself.

Right? So, Hubble is our internal sort of like data querying layer. And I'm going to tell it, hey, go find these queries that I usually use to track Kai adoption. And I want you to like create a dashboard for me. And this is really interesting. I really like this because some things that I've found uh most people really use AI for. When I talk about most people, I'm like, so I'm an engineer. I've been an engineer all my working life.

Uh but I'm super passionate about how we scale AI to like everyone. And not everyone is necessarily an engineer even if they're super technical uh at Strip. So something that people really have been using the EI for is creating dashboards that communicate a point and that's what I want to I want to show uh how we would be doing that using Kai today and there's a lot of stuff we could get into along the way >> and I see it it pulling for tools and skills.

So is this like part of the harness which is it's kind of like tuned to go find what it can use to solve this problem? >> Absolutely. So uh there are two things. Um we have of course we all have we all know what tools are and we all know skills are um tools basically being the things you can do and skills are like essentially a way to package up relevant tools so that we can find them easier. So the thing that you see Kai doing immediately is that it's going to find the skill that says ask data and that's a skill that's going to go figure out how to answer any stripe data and uh SQL related question using our standard tools.

Using that skill which it's loaded up it's got access to a bunch of like internal uh tools and it's going to use those tools and go ahead and and do more things right. Um so the tools are a combination of a bunch of different things. There are tools that we set up in the harness. Example discovering a skill is itself a kind of tool that we give the harness, right? But we also give it things like there's a sandbox that's secure for your session where you can do grip and stuff because again unlike our typical agent, this isn't running on your laptop is running in the cloud.

And so how do we make sure that your session cla and my session don't like eat each other? So we set up the secure sandbox and give the harness tools to interact with the sandbox. Uh tools that can securely send data to it. Tools that the sandbox can you the agent can use to search things and do things on the sandbox and tools that can like get data out of the sandbox. Right? So that's a bunch of tools. Bunch of tools all over the place.

Um but the this one is pretty cool because again we have a sandbox. You don't even need to know that the sandbox exists, but the agent is going to be writing uh some sort of script on your behalf. I don't even know what it does half the time. Uh I know it's secure, but it's going to go in. It's it's found the data. >> It's Yeah, it looks right. I I told it not to make any mistakes. >> So, [laughter] um so that's what these tools and skills are doing. >> Yeah.

I I I have a kind of separate question while this is running specifically about making great data agents because I took talk to a lot of companies and almost universally the first internal agent they build is specifically for this use case. It is like data querying dashboard visualization agent. And so I'm curious in Hubble, which I think you said is like kind of your data store query engine, were there any things that high level you had to do to make Hubble agent ready?

Um, oh yeah, because I see like query metadata and, you know, ask data and these skills. I'm just curious like if you were building an agent and you needed to ready the data warehouse and the query lay layer, what are a couple key things that you think are super important for folks to think about? 100%. Uh that is such a great question. So uh like I said, luckily at Stripe, we care about our data so much that uh we've invested a lot into both the the data querying layer.

Uh we use we use Trino as our uh data sort of querying layer and our warehouse in that perspective. Um we've invested a lot into making that super resilient, right? And those investments have helped agents like slam it like crazy and not bring it down, right? uh we've invested a lot our uh data um the data platform side of things we've invested a lot in a catalog of data and taring of data. So we have access to schema uh that can quickly tell us oh these are the relevant data sets that you might want to find and use and how would you use it right but even there are higher level investments as well there's a blessed analytics layer where like the really key metrics go in right and like there's a tiering system where you you there's an analytics layer if you fail that you go look at all the standard data uh dashboards that we have and you use the queries from there and if you fail that then you use the data catalog and search through for the high quality data sets and figure out how to use it.

Agents are incredibly good at figuring this out. However, the key part and you asked about the ask data skill. The key part is we have uh some really smart data scientists as well who sort of said, hey, this is a this is probably the right way that most data queries should be handled. And what the skill does if we dig into the cast data skill itself what it's going to be saying here is route to direct artifacts first use the uh use the analytics layer first and if that fails and fall back and fall back and fall back and until you like hit you actually hit the data catalog directly right so these investments were made for humans but have held up really well for agents because turns out the reasoning through it you agents have the same problem they they can answer the question, but they have no idea if it was the right query or the right table.

Uh, and these investments have paid off in helping helping guard that. >> I want people that are listening to hear a couple things and uh, you know, I'm going to make the Stripe team blush. I say this specifically about stripe a lot which is I think one of the reasons why stripe has been able to benefit so much from AI is prior to AI there's been a commitment to developer experience developer platform data platform analytics layers like all these things that made humans really efficient at the company preai are foundational investments that now give you extreme leverage when you throw agents at it And so, you know, when people ask me like, "Claire, what can I do to ship more product with AI?" They think I'm going to say something about product development, and I say, "Double the size of your DevX team, double the size of your data team." Like, work on platform investments.

Good for humans, good for agents. Um, and that's what we'll will let you run. The other thing I you said and I don't want people to miss because I love this idea of this like three layer um triage that a data agent can go through and that's really smart like find existing reports please then use the analytics layer to find the right query then if like you really have to fall down to the data catalog and write your own query.

The thing that I also heard you say is your data warehouse has to be very resilient to high volume queries because when in doubt an agent will just brute force it. And so, um, again, this is like infrastructure hardening investment, performance investment, not sexy, not what people are thinking about when you're building these data agents, but actually allow agents to do a really effective job because you don't worry about like, you know, turning over your your data warehouse because a agent is hammering it 100%.

And like everything that you said makes so it's it's resonates so much with uh with all of that. my my uh personal history at Stripe has actually been on each of the kind of teams that you referenced. So I'm like, "Yes, someone gets it." [laughter] Uh so this is great. Um the the thing about resilience um agents are very creative at bringing your intro down. What can I say? They're like, it's almost like all these scripts that they were trained on, just teach them to be script kitties or something, right? the the the thing that um we really did well is thinking about agentic identity like we haven't solved this yet, right?

But thinking about how do we say that you know this is the this is an agent and this is what it's trying to do like what is a use case it's trying to use to uh as it goes around doing its thing in our infrastructure and using that as a way to think about priorities and load shedding and all of that good stuff. Again, not super sexy, very like deep infra stuff, but the same principles apply. >> It turns out that agents just like dial up all your failure modes like it's just it's it's just ex it just multiplies the amplitude of problems you can get, right?

And uh the investments I I wouldn't claim that we did not have any issues. We definitely had a bunch of issues where when we started doing this like there were agents that went rogue. uh um there were agents that you know uh may have almost taken down core systems uh but we caught the caught it in time and uh and now we've hardened those systems as well. I >> I love it. Okay, so we've uh yapped while Kai ran. Let's show what Kai actually generated using these skills and tools in Sandbox. >> Yeah, of course.

Um so here's what you see. you see that uh you know Kai adoption is looking good and uh this is something I'm personally super happy about like pretty much everyone at strike uses Kai um like 86 plus% of the company now so really AI for everyone uh which is how we started out u process and you see this ramp that's gone from a uh fairly low number I I think if we had done this a couple of weeks ago it would have been in the hundreds up to a very high number so happy to talk more if you're interested if uh viewers are interested into how we uh manage that.

But okay, we have a dashboard. Dashboard looks good. It also looks like vaguely stripey. Uh so I need to go back and see how the agent figured out that it needs to make things blurable. So uh I got to go figure that out. But and it has a bunch of things here. It's an interactive dashboard and has links to a bunch of things, right? That's fine. This is great. We can already see how this can be useful for like I now generate a dashboard every meeting I go to because it's so easy and it helps me drive uh drive the meeting a lot better.

But the real power here starts to come in when you talk about multi-turn conversations, right? So great, you have a dashboard. Awesome. But let's do something more. Let's sort of like get K to uh iterate on this for us, right? So, hey, I love this dashboard, but let's do some more here and uh use this query get a breakdown yada yada yada. And it's going to do some interesting things here. So, I'm going to kick this off, but I'm going to talk through what I'm doing, right?

A the dashboard isn't like is the artifact isn't like created and it's not fire and forget, right? We give a chance for people to do deep work by iterating on their artifacts and that's really powerful. Uh it's better for token efficiency. You don't want to be throwing away a HTML dashboard every every turn. But it's also really moving into this idea where the agent and you are collaborating on a task, right? And uh a we have turns that are like super deep like hundreds of turns uh over multiple weeks.

So the idea here is you have sort of like a somewhat like a pretty smart collaborator who has some artifacts and you can iterate with them on it. I'm going to add this query and I'm going to do some really interesting things and this is something that I I think it's worth getting into. I'm telling it, okay, it's not just pulling the data. It's not about pulling the data and displaying it. I want you to do things with the data.

I want you to like munchge the data in some way or form so I get what I want. And the reason why I'm touching upon this is a lot of the data sort of things that people want to do end up being last mile data. You think about people's workflows. It's so different. It's so hard to build a dashboard for everyone to do every part of their job and then then you have like a gazillion dashboards and how do you manage them? You can't keep the right dashboards at the right level of quality.

Using an AI like Kai to do this means that you can create like light apps almost like the whole like the lovable style uh thing where people are creating apps to just hyper optimize for their workflow. And the fact that they have a sandbox that anybody regardless whether they're an engineer or not can get the agent to write code for them and do whatever the heck they want with the data. It's really powerful. And I'm pretty sure that again as expected it's gone in it sort of like said okay here's the actual data and I want you to go do some summation somewhere to do the other tab and um out it came.

So uh super interesting and if I open up the updated dashboard it's the exact same dashboard and you should now see this really cool little uh segment below that. So I I could I could keep yapping about this dashboard. >> [laughter] >> I love the fact that our marketing team is like a 100% allin, [laughter] right? >> They need it, right? I don't know a single marketing person that doesn't either want some sort of app built or some sort of dashboard.

So, I think you have product market fit. That leads me to my next question, which is how do you roll out I'm just curious kind of, you know, inside inside the doors of Stripe. How do you roll something like this out? Is it really organic adoption? How did it get built? How did it get shared to the team? Was this like 20 engineers team like h how did this come to be? >> Definitely not 20 engineers. [laughter] Uh Stripe uh we we run fairly lean and very nimble and very fast.

Uh we built this super quick. It it it's a really interesting case because um uh so we had like say me and my colleague we had the idea um um I was moonlighting as an as an engineering individual individual contributor again trying to get this out the door and so it took us like one and a half people over two weeks to get Vzero out the door and something we realized was a lot of the questions became the answers became apparent once we could show people something.

It was very hard to tell people why something like this is is required in a world where you had the coding agents around and they could be like super powerful. But the moment we got that V 0ero out, super inexpensive, one and a half engineers for like two weeks, right? V 0 out. And then we moved into a pilot stage where we started seeing a lot of um um interest from primarily we had a great collaborator uh Ilia on the GTM team um who builds like AI for GTM, right? and they were like super interested in this because um like marketers are go to market uh function at stripe like they are extremely like they're looking for whatever they can do to to reach more people to reach them more uh in the right manner and so on.

So they they saw a lot of adoption from them and um that's how it kicked off. It went into this pilot stage. We still had about 200 to 300 users at this point. We had like two and a half three people working on it. Um and this was the next like month or so things really ramped uh once we did like a companywide demo saying that hey we built this thing uh we invited to use it and it just clicked for everyone and people started that's when you see the really steep ramp up somewhere here and everyone started using it even then the team itself I wouldn't say is humongous like yeah 10,000 plus people use it every week but the core team that manages the experience is still like less than 10 people.

It's uh and we have a lot of other things we have going on uh with those 10 people as well. Um and the things that let us build it, of course, coding agents and the in the productivity that they've given us and our devro team are incredible. We've spoken about minions on the show before. We have incredible tools at Stride to get more from uh from who we have. Uh and we have all this infrastructure that you referenced and all that has helped us.

So, I would I would say it's less than 10 people, but I also want to give credit where it's due. Like, there are a lot of people helping those 10 people do what they can do. >> I love it. This episode is brought to you by Hyper Agent, the platform for deploying always on agents that actually run your business. With Hyper Agent, you build agents in the cloud and deploy them where your work already happens, like Slack, Telegram, or email.

[music] An agent will scan your inbox and draft replies to vendor follow-ups. Another monitors [music] competitors and spins up rich ad kits and landing pages. A third notices a deal going cold in Salesforce and writes the save email with full account context. These aren't chat bots waiting for a perfect prompt. They're proactive, [music] learning your preferences, retaining your playbooks, and getting better with every run.

One user built four agents to run an outbound sales pipeline, prospecting, outreach, follow-ups, CRM updates, all in a single afternoon. No local setup, no VPS bills, no fragile permissions on your laptop. Just powerful agents with full control over skills, tools, and guardrails. How I AI listeners get $100 in free inference to start building. Claim yours at hyperagent.com/howi. What else? maybe one or two other things that you think are worth pointing out in Kai um that you think make it pretty unique or at least you know fun and easy to work with. >> I'm going to do two things.

I'm going to show skills and I'm going to show projects, right? So when I come to say skills, we have this thing. It's great. Everyone loves the dashboard, but I don't want to be creating this dashboard and paying a bunch of tokens and uh time every time. So the thing that I uh think Kai did really well and one reason for it product market fit was I can create a skill that basically takes what I've done in the session and packages it up so that it can become a loadbearing repeatable workflow right and that's when the AI goes from here's something I'm just like iterating with on the side like a chat interface to here's something I can trust to run my to sort of like run my business or run my workflows or help me um help me do that uh a little bit after you're going to see this kick off this like skill creator skill.

It's a skill that the harness has that's going to go in and create like take all the things it's learned from this session from its interaction with me and package that up nicely into something that I can just like um uh pull up at any time and we'll we'll talk a little bit more about um how we do the skill um uh retrieval. I think that's a really cool part of the system as well. But but while that's cooking, right, uh let's actually look at this other thing I I love about how we built Kai, which is this notion of projects, right?

So the thing about projects is there's so much stuff happening at St. We have like 2,000 skills, right? Projects are this really nice packaging mechanism where we can draw a boundary around those skills and say these are what most people who are doing this workflow must be using. We have projects that are created for uh projects like shortlived things. They have projects created for teams like the people team has a super secure version of Kai in a different project that's backed by a totally secure back end and stuff.

The thing about this is again it lets one person or a few people who are DR of the space to uh figure out how to um get the agent to perform well for everyone. The really interesting thing I have on projects there's this thing called settings that like I said you can use it uh you can use a custom agent to power your project. It doesn't have to be archive which uh which is pretty good and general purpose. But let's say you have something really bespoke.

You can use all the same features we have but just backed by a different API in the back end and a different harness right. So really again when you talk about uh AI at the enterprise there's going to be hetereroentity there's it's going to be a lot of different cases and building this in layers so we can give maximum leverage and customizability. One of the cool things about projects that can be very concrete for people is the idea of tool policies.

Now, I mentioned the people team. Let's say you're a person on the HR team who's dealing with a bunch of sensitive information, right? You really don't want the agent to sort of go rogue and put that sensitive data into some public Google document that all stripes can access. That seems like a accident waiting to happen. We don't like that. But you also don't want to tell them, oh, you can't use any tools because your workloads are too sensitive.

Right? So what projects let us do is to say for this workflow, I'm going to set up a tool policy that says in this case I set the run Hubble uh tool so that I don't inadvertently uh put some uh confidential information into the demo. Right? But uh we could test this out with some other tool and it's going to uh kick off what is a um is a human in the loop workflow. So create a calendar invite for me and Wong tomorrow at 11:00 a.m.

Pacific, right? And I set this up ahead of time just to show what a human in the loop flow would look like. Uh but you can you can see it extends to uh any other kind of tool. What this is going to do is going to tell me hey >> should be familiar to most people uh who've used like cursor or uh or the other uh large you know big products out there but I want to do this right the interesting part and why this isn't the fun part the fun part is what I showed before which is that someone who is the DRR of a space can decide that certain tools are kind of sensitive for the workloads that these people are going to be using.

So we need a human in the loop to confirm if that action can be taken by the agent. Agents are really good. They're very creative. So we got to put some restrictions on them so they don't go rogue, right? And projects help us decide. I also don't want this to be happening for every person at Stripe. That would be kind of frictionful. So projects are again drawing a boundary around it. >> Yeah, I I love this because I think a lot of the existing tools let you maybe configure some of this at the individual level, but then it applies to every session. it's not contexted to what you're working on and you can't share that permission set across different users.

And so what I think is interesting about Kai is the like permission and context boundaries are very purpose-built for how your company works on things. Um, and I think, you know, when people are asking themselves either, should I build something myself and does that make a lot of sense or do I need to pluck something off the shelf for my enterprise use case? Again, you need to ask yourself, how much appropriate or inappropriate friction will this put in everybody's day-to-day work?

Because at the end of the day, what you want to do is make everybody's life easier without causing chaos or trouble. And you want the management requirements to go down really low, right? You don't want everybody to have to think every task like, do I need to turn on this connector, off this connector, connect to this data? Um, and so I I do think one of the benefits right now of of teams building their own thing is they can really think about bespoke agents for bespoke use cases, but kind of hide all that complexity from the end employee, the end teammate, um, and just let them get to get to work.

That that's super insightful because as you were speaking about bespoke agents and I showed a little bit of this earlier, Kai looks like a single product. It really isn't. It's like the icing on top of a multi-layer cake and each of those layers can be like customized to work at the enterprise. Right? So, uh 100% agree that the notion of both customization but lowering the cost of management, the cost of ownership and just the friction.

If you put too much friction in front of people, they're just going to do unsafe things because that's how humans are, right? We we we don't if you if I showed you this every single session for every single tool, eventually you're going to press the wrong button, right? Uh so really thinking through that is a big part of what we're trying to do here. So that's projects and why I love projects as a unit of governance.

Let's hop back really quickly to the skill builder flow that I spoke about. Again, we made that dashboard. We want to make this something that I can uh that I can reuse, right? And it's not just me. It could be my entire team. Why do we have to like keep things uh close to ourselves, right? And so I can now go KA did a skill for me. It's like a standard open spec skill that you can use on any of your uh harnesses, right?

But for the purpose of this, I'm going to go click this button and Kai is going to say, "Okay, what do you want me to do?" Uh I'm going to go ahead. I could choose to push it push it to an area. And we'll talk a little bit about area skills. But for now, I want to keep it private to myself, right? And of course, this is like nothing special about me here. Any user of Kai can do this. And that's why we have a lot of skills uh now that are making people more uh effective.

So, it's filled in the description for me. It's filled in like when Kai could use it. So, this is important. We'll come back to this in like a second, right? And I'm just going to go ahead and export draft. Oh man, I already created one. [laughter] Um, again, uh, this is what happens if you prepare too well for devils. >> This is when you when you do it live. I mean, we we believe you. I think what what you're showing here.

Great. You have kind of a skill creator skill or tool. You have a specific spec that you're using to ensure that it's both written well generally for agents, but also written well very specifically for the Kai harness. Um, and then I love this idea of a draft >> kind of skill editor, >> um, that you can test and edit and optimize and manage. It's quite nice. Um, I know people just love fussing around in Markdown and, you know, Python files, but just a little quality of life UI here can go a long way. >> 100%.

Um, we really invested a lot in making this feel like a a little bit like an IDE so that everyone can get access to that quality of life improvement. Um, go in back here for just a second. Now that I did this, I can go and this is the magical part. This is the part that I'm really excited about. [laughter] Um, and the the team has like really kicked us here. So, give me the latest PI adoption dashboard. Right, that's all I'm going to say.

And this is the part where the magic of Kai really shines. Now, when you're in a coding agent, you can see what it's picked up. It's picked up the skill that we just like literally just created, right? And the reason why this is interesting is when I said all the way at the top of this that you can just go in and start using Kai and it knows what to do. This is how it knows. Looks like my uh my my tool policies are too uh are too secure, right?

So, um now the thing that they've done is in a when you're a coding agent, right? you're in a re repository, you're in a folder, you have this natural structure to what you're trying to do, right? And so you can pick up the uh the the the skills in the hierarchy of where you're working and you get the right set of skills required to do your job. When you're at an enterprise and you're starting to work, you don't know like there is no hierarchy of what you're going to do.

You're frequently trying to hit like five different systems. And so a large part of the investments I've done and what we've um what we've managed to give to Stripe is the ability to package skills and retrieve them and we do this like really rigorous flow of knowing when the right skills are being invoked so that Kai can perform at a high level right and the number of skills that we have we got to do some pretty uh interesting things to ensure that we keep those skills in tip-top shape.

We've also built out this interesting thing where it's not enough to enable people to build a bunch of skills. How do you make sure that they actually know what it's doing and how you keep them in in top shape? And that's the other thing that we're really investing in this automatic platformdriven like suggestions for how to improve your skills. So that especially since we let anybody publish a shared skill, right, that anybody else can pick up, it's become really important to ensure that we can give people the tools to keep those in tip-top shape.

Again, something that you probably don't worry too much about if you're just using AI for yourself, but the moment you introduce like sharing and the enterprise, uh, quality, governance, policies all become something really important and they're usually something pretty specific to the company you're working at. And and Stripe is no different. You know, the only other thing that I've seen here that I'm curious, maybe you have, but you haven't shown, is we see a lot of folks um that are building these internal harnesses do skill and tool telemetry and observability and see where like tool calls are failing a lot.

So, they can auto eval. And then they also have a deprecation policy for skills. So, if skills have not been invoked >> for like 30 days, you get a little notice and it's like, hey, you haven't used a skill. if it's dead, maybe we archive it. And if they don't get a response, it goes into like um deprecation status and then they delete it two weeks later just to like prune all this stuff that's happening. And so I think this enterprise level maintenance of the skill library is really important not just from a quality perspective which is what we see here with the evals but honestly from a quantity perspective like just is any of this useful anymore >> 100%.

And uh it's I almost think you can't separate quality and quantity when it comes to uh these systems because you know context is everything. the more unrelated context you throw into the AI, the less good your uh results become. So quantity is almost a uh facet of quality. Uh we've uh we started out with like a bunch of different uh skills and we do have telemetry um on we have let's say 50 skills that are used like hammered every day across the company.

We have this long tail of 100 to 150 other skills that are used by subcategories uh you know parts of the orc chain and we have a bunch of tools that are used by two or three people right we need to respect that all of these exist there are teams there are three member squads doing some really bespoke thing and they want to share between themselves but then again what is the telemetry we have and part of the process I'm showing here is this is the userfacing part of it but part of the process is telling us as harness owners these are the kind of skills you probably want to promote up into like a general workflow and these are the skills that you want to like delegate out and move out of the general workflow because it's just taking up context right so uh definitely uh unfortunately don't have something cool I can show around that but it's it is something that there's an ETL pipeline happening somewhere that's doing this >> amazing well I just want to recap for folks because this has been awesome just highle things about Kai personalized context um for people. org chart awareness um you know tuned tools a sandbox that you can put data in a sandbox you can pull data out shared artifacts >> projects which I feel like if you missed that part rewind go back to it because projects are not just how you organize chats across a team but how you give a specific space whether that's a team or an initiative access to tools access to data permissions and controls including what requires a human in in the loop.

Um, a skill builder skill, but not just a skill builder skill, a skills platform for the company that allows you to build skills, edit skills, eval skills, share skills. Um, and then you know AI that just just works for the things that matter, including data analysis, which benefits not just from this tuned harness and skills around it, but investment from infrastructure all the way up the stack on a great agent ready data layer.

And it took one and a half agents or one and a half humans, sorry, probably a million agents, >> many more agents, >> a couple weeks to get V1 going. and now is serving 10,000 stripes with less than 10 people plus a bunch of great infrastructure that you've been investing pre and post AI. That's it. That's all >> that that's all. [laughter] Not not much not much to it. Um but yeah, it's it's uh I think that was a fantastic summary.

A lot of good stuff. Um I think it's important also I would be remiss if I didn't say it sounds like we figured this all out. We absolutely haven't. like we are we are very cognizant that we are in the earliest parts of this journey and um we're hoping that like you said the strong foundations we have help us iterate and move forward with the with the times. Um but yeah, it's been a fantastic journey so far and uh maybe we'll be back in a year showing you something completely different because that's how quickly the space moves. >> I I really hope I hope I hope sooner than a year.

Well, before we get out of here, let's do two lightning round questions. Um, my first one is let's just put Kai aside for a minute. Let's put a put put aside Stripe Blurle. What are personal AI things that are fun that you're doing or that you're excited about as an engineer? Um, you know, when you shut, you know, as you say, we like shut the work laptop and open the fun laptop on the weekends. Um, what are you excited about?

What's cool? I I'm super boring, but So, I'm going to like say something less boring first, which is uh it's helped me like not sound super dumb to my to my six-year-old who's right at the stage where he's asking me all these complex questions about like exoplanets and um and like galaxies far away and I'm like on the side, you know, Gemini on my phone like, "Hey, like can you tell me what's happening?" And then I act like I know the answer.

So it's helped me, you know, keep up to his model of dad knows everything. So that that's that's good. >> Perfect. >> Um, but on the work front, honestly, like the stuff that I'm not doing when I'm building Kai, it's like my I have this whole workflow now, which is around using the AI to make sure that I don't miss things. It's so boring, but it's so good. Uh, we didn't get to show you Kai schedules, but I basically use Kai as like my personal assistant.

It just tells me things that I'm supposed to be doing and so it's so basic I I almost I'm almost embarrassed to say it out loud but it's it's been the biggest life hack. Um just not having to keep it all in the in in the brain has been awesome. >> Yeah. I what I tell people is we think a lot about how to put agents to work. I want the agents to put me to work. I want them to say Claire, please fill out this form. Claire, please do this thing you said you were going to do.

And so I think it's a it's a give and get relationship. And I love that. And I also have many children who ask me really existential questions about the universe and about dinosaurs and about history. And I agree intelligence on demand helps us keep our superiority in that par parental child relationship. >> Yes. Yes. For a few more years at least until they figure out what we're all doing. >> So last question. When Kai Well, maybe not Kai, maybe you're very sweet to Kai, but when AI is not listening, what do you do?

How do you prompt? Are you a yeller? I I don't know. I feel like I'm maybe I'm just subconsciously afraid of uh what it's going to do to me when it figures out, you know, knows where I live or something. But I'm very nice to the AI. I just say, "Hey, that's not what I wanted. >> Here, I'm going to say it again." And then maybe I'll like all caps it, but I I don't know. I I don't like shouting an AI. It feels uh it feels wrong.

It's almost like I'm shouting at the people who built the AI. So, um [laughter] maybe. But I I just insist. I just uh I say yell harder to my team, but I actually end up just like saying a lot of please, if you will, read the thing I said better. But is it does mess up and it's very frustrating. >> Like like many of our guests, you gentle parent the AI, which is >> this is true. >> I I know I I know you can do better.

I believe in you. [laughter] >> I'm I'm not mad. I'm disappointed. >> I'm disappointed. I'm just disappointed, Kai. Like, you should have done better. >> I love this. Well, this has been super helpful and interesting for me. It's given me so many ideas about just my own use of AI and how I talk to people in enterprises about their use of AI. Where can we find you and how can we be helpful to you and the strength team?

Well, uh, I'm on, um, I'm on LinkedIn and, uh, I'm happy to connect with anyone who's like super interested, um, in learning more about what we built here and how we think about, you know, uh, scaling AI for the enterprise. Um, and Stripe is always hiding. We are always on the lookout for people who want to, you know, join this crazy band of people trying to build amazing things for for the world. So, um, please look out on the Stripe careers page.

Um, but otherwise, um, my email is, um, um, shared.com and I'm happy to, um, happy to engage with anybody who has questions about anything we covered today. Um, but yeah, keep those questions coming. >> Awesome. Well, thank you to you and thank you to the Stripe team for being so generous with all the things that you shared with the audience. We really appreciate it and thanks for joining. How I AI. >> Uh, thank you for having me and this was uh, this was uh, fun.

Uh, I don't know if I mentioned you're a minor celebrity uh, on the team. So, uh, I now have some reflected glory and, uh, yeah, it has been amazing time. Thank you so much. >> Thank you. Thanks so much for watching. If you enjoyed this show, please like and subscribe here on YouTube, or even better, leave us a comment with your thoughts. You can also find this podcast on Apple Podcasts, Spotify, or your favorite podcast app.

Please consider leaving us a rating and review, which will help others find the show. You can see all our episodes and learn more about the show at howiaipod.com. [music] See you next time.