Redpoint's Sai Senthilkumar: Cursor hit $500M ARR in under a year — AI dev tools are becoming winner-take-most

what's happening here But it is seemingly clear that um the uh founders, the team and the investors in Scale AI will be getting um paid out I would imagine proada based on that $15 billion um mark. And the and the thing that's wild here is I mean I so it's an up round for everyone.

Everyone gets everyone gets cashed out gets a ton of money. Alex gets to go work in like a hyperscaler in super intelligence and actually like take a run at some really serious like high technology. Yeah.

The thing that was ultimately surprising so Scale AI has more than $900 million in cash on its balance sheet at the end of last year according to financial information shared with prospective investors. Uh they've raised more than 1. 5 billion from investors.

So they only burned 400 five 600 million of that like they kept more than half of it. Uh the company made uh 870 million in revenue last year and expected more than two billion this year and uh it lost about 150 million. So I mean doing doing pretty well.

I mean there is a world where where like scale AI as a company even though they're not wholly owned by Meta, they are independent and they have $900 million war chest and they can go build a great business and do a bunch of stuff there and then uh that the relationship with Meta can be a fantastic growth engine.

Yeah, this the the only reason this was surprising to me, well, I guess one, it's um scale AI was, you know, if you're if you're trying to build a a new research lab, they're probably going to need to still acquire a lot of other top talent.

You could imagine Meta going and and trying to bring in into you know, people from Anthropic or OpenAI into this new organization. Um, but uh it felt like scale. I I I do wonder if if Scale, you know, Scale's team could go back and and see Circle's IPO and the success of that. Yeah.

If they would have still done this done this deal because Scale AI could have ripped in the public markets. Yeah. Circle IPO showed that the public markets are very receptive to pure play technology bets that that have great narrative around them. Yes.

and and I could easily seeing what circle has traded at I could see scale trading well beyond um uh you know the this valuation they're getting from meta. I agree with you. The question is how does the liquidity picture what does the liquidity picture look like in that scenario?

They wouldn't the in insiders wouldn't have been able to unload 15 billion 14 15 billion like does that does that actually play out?

Um because yes, yes, the the stock could have traded up to a 100red billion market cap, but can you actually if you're an investor, can you actually get liquidity without crashing the stock if you're putting 15 billion of downward pressure? Uh so yeah, it's a lot big question.

We'll have to ultimately huge pickup for for Zuck. I think Alex, you know, will will continue to be um you know, execute uh very well and I can imagine he'll be a huge part of recruiting. Yeah. Uh he's done a fantastic job. He's like been everything.

He's been on everything from like the Theo Vaughn podcast to testifying in front of Capitol Hill and lawmakers. He's got a ton of range. He's very able to storytell and do creative deals which increasingly are extremely valuable in the world where the AI paradigms like the secrets are known.

It's about, okay, how can you actually marshall the capital and and the will to build the hypers scale data center to actually go and and get Jensen to fork over 100,000 H200s on on schedule? Um, I'm sure there's a ton of different uh things in the works. Yeah, it's interesting for Meta, too.

Even if they're able to leverage the scale team uh to simply get better at ads, they can make back the the the 15 billion. Uh oh, yeah. in there's a ton of different ways this works out.

Uh and I mean they clearly need a lot more reinforcement learning training data for to improve llama 4 because they need a they need a reasoning model um if they're going to stay in the game.

That's clearly the next paradigm and I mean like a big part of what we learned about uh from semi- analysis this week was that verifiable rewards are really important in reinforcement learning. Scale AI pioneered that humanity's last exam.

they've been in like the eval game defining the rewards that you would work against in the reinforcement learning paradigm. And so, uh, there's a whole bunch of ways that that plugging Alex and the scale team in makes a ton of sense. Anyway, really quickly, let me tell you about Adio customer relationship ma magic.

Adio is the AI native CRM that builds, scales, and grows your company to the next level. Let's give it up. Go check it out. And we have Let's give it up for ADO. Red Red Point, our first guest, investors in audio. No, wait. Really? Yeah. What a coincidence. One hand watches the other. Welcome to the show.

How are you doing? Well, what's going on? Great, guys. Great to be on. Huge fan of what you guys have built. So, this is exciting. Thanks so much. Uh, would you mind kicking us off with a little bit of an introduction? For sure. My my name is Sa Lumar. I'm a partner at Redpoint Ventures.

We're a venture capital firm in the Bay. It's been around for over 25 years. We've invested in companies like Netflix, OpenAI, uh, Stripe, but we're known for our infrastructure software investing.

So, companies like Stripe, Twilio, Hashi Corp, Cribble, Chain Guard, and yeah, that's been our bread and butter and and and walk us through what's going on today. Yeah, so we're we're in San Francisco at AWS's Builder Loft. We've partnered with NASDAQ to actually move the ex their exchange to SF.

So, in about 30 minutes, we're going to ring the bell. Really about 150 people here. Yeah. It's it's like it's like that Deliant tweet like move move Silicon Valley to Miami. We just move Wall Street here. So, we're excited. It's exciting day for us. But yeah, this is our annual conference and we do it every year.

That's amazing. But break down. We got a hopefully you guys have we have a gong cam here. Hopefully you guys have a bell cam set up, you know. But um the gong might be better than the bell. I think I mean yeah SF SF is very gongcoded. Yeah. Yeah.

Um well yeah I I want to know about the top themes uh uh for for the conference the top tracks the top discussion points what are people debating um for sure what uh uh take us from kind of uh where the discourse is and where the frontier of that discourse is going. For sure.

You know, we we created this whole initiative a few years ago to call out the importance of infrastructure software as standalone category. It's it's it really is like the picks and shovels behind how software is consumed and delivered. It's this invisible piece of software that we just take for granted.

Um when you when you buy something on Amazon, for example, there's some application that we open, but it's powered in the back by all of these dev tools and databases.

there's some global server taking in billions of bits of information and you need to secure it from all of these DOS attacks which happens millions of times a day and we just don't know about that.

So I guess across the whole application delivery life cycle there are just a ton of different infra tooling and it powers virtually everything we do. So yeah we we take uh we we we do this conference every year to just shine a light on all the amazing innovation that's happening at the infrastructure layer.

So we we divide the world up into into four different subcategories.

Everything happening at the AI and model modeling layer, all of the data infrastructure tooling that needs to be built out around that, the cyber security vendors that need to protect that data and all of the developer tools that actually creates the software.

So we've been we've been doing it for uh this is our third year doing it and we also highlight basically the top 100 fastest growing private infrastructure companies. we all get them in a room. We have the C CEO of AWS, Matt Garmin, coming to chat today and the president of Gary of of YC, Gary Tan.

And so, yeah, it's a it's a fun day. I have this I have this thesis I want to bounce off of you and get your feedback on. It feels like DevTools infrastructure, it's less monopolistic. It's And maybe that makes it easier for maybe not VCs to make money, but uh entrepreneurs to make money.

you just don't get steamrololled as much.

Am I am I off there in in thinking that uh a lot of these dev tools markets are more oligopolistic by nature of the fact that hyperscalers are competitive and the different cloud platforms are competitive like what what is the current dynamic because I feel like the narrative is always like someone comes out with some infrastructure or dev tool it's kind of a boring company it flies under the radar for six years and it gets and then oh it's worth a couple billion dollars and then it's worth like 2030 billion no one knows about Exactly.

So, so, so we break down like what structurally is going on there um why is it harder to build uh these crazy crazy monopolies and but and then is that narrative about like base hits being easier or unicorns being easier is that even real? I actually know I I agree with that sentiment maybe a couple of years ago.

But if you look today, I mean look at cursor hit $500 million in ARR in in less than a year. And I guess like dev tools, it's always been this like very sequent like if you think of DevOps, it's this very sequential process, right?

It's like you have if you have your engineers on one side, then you have your your S sur folks on the other and then there are there are tools for the the developers, there are tools for the incident response engineers and there are like these distinct stop gaps that exist across the life cycle, right? Yep.

But with these code generation tools like cursor and windsurf and copilot like they all started on the left and they started in the IDE where these developers are actually coding and but you can see the writing on the wall and where they're going.

So now you you're you're consuming and writing all these applications in the IDEs, but they're probably going to extend, right? And they're going to start doing the code review stuff and then you're going to deploy the apps.

Like you can see it in Lovable, for example, with just like a simple query, you can go from that query to a full-blown production app that's live on a website. So like before you needed like 10 different designers and engineers to do all of that, but now with just one tool, you can do that.

So I think I I think yes it's it's not been as monopolistic in the past but I think that world is changing and and like the clear horses right now in my opinion are cursor and anthropic in the coding world. Interesting.

And and so yeah I mean it's just crazy like two years ago like people were saying LLM all they could do is just spit out code like you couldn't actually write real software and impact software development.

Q two years later it I think there was this interesting stat like computer engineering grads face double the un unemployment rate of art history majors and it's it's it's just crazy seems crazy um h how do you compare and contrast the adoption of of uh AI versus traditional cloud?

I I know that you guys have done some some research around you know cloud adoption uh you know more than a decade at this point versus um even just the the the growth rate on on the token side. Yeah. I mean it you know it's it's just incredible how how much faster the consumption is on the AI side.

I think a really good analogy or data point is if you look at the cost of EC2 like servers during the cloud the it got efficient like very very quickly and it allowed people to consume cloud software all across the board but when you look at the cost of inference which I think is the equivalent to EC2 and cloud but inference for AI it's it's it's literally dropping like a 100 times faster than than the cost of EC2 but at the same time application patients are being built and consumed 10 times higher than it was during cloud.

So it's like a thousandx more consumption at the end of the day.

So it's it's just it's just insane like what what's happening in the age of AI and and the markets itself like there's this massive services component that is going to be converted to product based software revenue because you can encapsulate all of that all of these workflows with software like LM are software at the end of the day and and you're you're taking these huge markets and you're making them available through software.

How are you thinking about competitive dynamics between sort of scaled infrastructure providers like data bricks and cloudflare and and companies in that category versus some of the upstart uh companies that are scaling rapidly but um you know still having to compete with these you know founder companies that are um you know basically uh they're not going to let uh they're not going to miss this sort of like platform shift.

Yeah. You know it's it's really interesting.

I I think the existing infrastructure vendors they can actually embed the AI products in their suite like take like take vector databases for example which I think is is an incredible market still right and there's like all these use cases but is there a vector database equivalent of data bricks yet like no actually what happened was that MongoDB extended vector search part of their as part of their suite so I think like but but like that's that's not to say like like for example like two years ago people would not have thought an ID like cursor would would be better than GitHub copilot.

So I think there are pockets of the infrastructure market where sometimes the incumbents are going to win and they can just leverage their existing technology to extend their applications but then there are like AI native wedges like the IDE for example where there's tremendous opportunity for a couple of 22 year olds to come and create a 10 billion business in in two years.

Yeah. Let's stay on cursor windserf the new uh the new dev tools like IDE coding AI coding market. Um what's I I'm I'm I've kind of been there's a little bit of a horse race going on. You know, you're either a cursor guy or a wind surf guy or you're an investor in one.

Um, but as I talk to more and more of the founders, I just increasingly become convinced that this is such a growing market and it remains potentially oligopolistic that they might kind of all wind up winning in some way or at least returning investor capital and and making the founders uh fantastically wealthy and successful.

Um, how much of that narrative do you think is real? Uh, we talked to uh Scott Woo over cognition. We were like, "Oh, like uh OpenAI just launched Kodak like are you cooked? " and he's like, "Well, we grew 40% last month. " And so it seemed like very much not so.

And we talked to another person who was saying that uh GitHub Copilot not the trendy tool at all at a $500 million run rate. And so that in and of itself could be a public company I if it if it was spun out.

And so it feels like when when we're talking about doing something so net new, something so additive to companies workflows, there's just so much opportunity that it kind of just is like a rising tide lifts all boats. But uh what do you think about that narrative generally?

Yeah, I mean if if you just look at the application with the strongest product market fit within the AI world, it's coding and and like the market itself, there are tens of millions of developers. If you just assume moderate assumptions in what you pay them, it's like a $1. 6 trillion market in spent.

It's it's like it's absolutely massive. It's like the mother of all markets. And I think it's really interesting. There there are companies going after certain pockets. So you have the code the code generators living in your IDE and that's where the developers are today, right?

It's it's where the actual development happens. So cursor it was just ma magic in a bottle. It's tap tap tap and all of a sudden you have a full-blown production app and you save yourself countless hours in in in development.

Um, but when you when you can see where the world is moving, like all of these processes think about humans at the center, but it's like very very quickly we're think it's going to shift to having agents at the center. And so companies like Cognition and Factory, they're not going to exist in your IDE.

Like you're just going to go tell it. You're going to go tell an agent to go do something and it's going to come back and have a pull request and you're just going to approve.

So I think I think there will always be a need for software developers, but I think they're they're going to move increasingly from actually coding to being kind of these like orchestrators of of a product. And I think like product managers, for example, are now enabled with these tools.

I think at the end of the day, like I don't I don't think it's going to Yes, there there's like some developers that might lose their jobs, but you're you're actually just making their their jobs a lot easier and there's just going to be a ton more software. There's going to be more software.

If you're not Jevans paradox pled now, you never will lost. reacting to the news today uh the scale uh AI meta deal.

How how do you expect the data labeling market to evolve now that now that in in in some ways one of the key players in that space is has other it will naturally have other priorities if you know half their their company is is owned by Meta who has um you know their their own goals around super intelligence. Yeah.

Look, if if you're the model providers, there are like two use cases that are just so clear that and and hair on fire. One, that's like inference. It's actually running these models. Two, it's data labeling.

How do you actually structure your data and clean your data and feed it into these models and such that you can actually advance uh the LLM? And so, I think it's I think it's a brilliant acquisition.

And I think whatever the sticker price is, like it's we're we're going to look back, it's going to be like the Instagram deal. Like you need accurate classified data to feed these large language models and for a variety of use cases.

There's so many data labeling businesses that we see like pockets in in hiring and and whatnot and and and coding and and like they're all screaming out of the gates like they go to like zero to 20 in like two months. And so I think I I I think it it just makes a ton of sense.

And I I think it's I think it's a brilliant acquisition. I want to get your reaction to WWDC. Felt like Apple was kind of pulling back from some of the territory that they'd overexpanded into in artificial intelligence. Obviously, it's a developer conference.

Do you think that there's opportunities for startups to uh build new companies even around servicing the nent AI for Apple environment? You could imagine there is like there is no cursor for Xcode. There's probably a way to solve that problem even just anthropic and their deal with Xcode. They're working on it. Yeah.

But but but there's probably some way to even advance that. And then also just serving up models into you know iOS APIs that are like more tailored. Like there's been a couple companies that developed backends that were specifically tuned for iOS apps in the past. Some of those wound up getting sold.

So, how are you thinking about opportunities coming out of WDC or or just general reactions? Maybe you don't even follow it at all.

You know, it's it's interesting when you look at when you look at like past platform shifts like the cloud, you you you first need like the workdays and service nows and Salesforcees running at massive scale like pushing the limits of of like cloud rails before you can before you can really understand what kind of innovation needs to happen at the infrastructure layer.

So, you know, our guess here with Apple and all of these apps that are built like you need the AI native versions of these apps that will start like hitting the limits of the underlying models before you can really understand. So, I for sure it's going to happen but I think it's going to take some time.

Uh what is uh what's the discussion uh or what do you expect the discussion to be around agents? There's great narratives around agents right now.

There's individual agent products that people use whether it's coding agents or or or sort of research agents but uh despite you know Salesforce you know marketing uh whatever their agent cloud agent force things like that it doesn't feel like uh they maybe have broad adoption yet but a lot of promise so so how how do you think about kind of agent adoption um given uh you know there's people I think at the event from on the browser infrastructure side, agent, devops, all that kind of thing.

Yeah, look, I mean, it's it's clear it's next paradigm. I think we're just catching it in a point in time where they're they're stumbling agents, but like again, there are some clear use cases where it it's it's clearly working today.

Like if like cognitions agents, for example, there are low-level development tasks where like that is fully automated and the largest of companies know that, but like no one else really knows that yet. And it's it's it's it's mind-blowing what's happening. Same thing with customer support.

you have you have companies like Sierra and Decagon and that's just a use case that makes a ton of sense for agents to operate.

So I think as the agent infrastructure improves there are things like memory and context budgeting there all a lot of these like hairy infrastructure problems that need to be solved but ultimately also the agents are going to improve dramatically like from a year ago.

You know agents have just gotten so much better a year from now they're going to be even better. So as agents improve all of the infrastructure around it improves like memory like I think very very soon there will be an agent for every industry. Mhm.

Do you think it do you think that uh it it ends up looking something like regular SAS or from a from a from a user standpoint in terms of you have sort of a dashboard or do you think there's potentially uh totally novel paradigms?

Yeah, you know, I I think UI is is just a big problem or or just an opportunity for these companies to kind of rethink how you surface these product and extend these products to your end consumers.

If you look at agents running in the developing world, in development world, they're just running in your CLI in the background. Is that the best way to do it? I don't know. Maybe voice is the best way to interact with them in the future. So, I think I think that UI is still being re reimagined right now.

I'm I'm guessing what we have right now will be thrown out the window and and that these abstraction layers will be very very different in in a few years. Yeah, the voice thing is interesting. That was actually in that original vibe coding post from Andre Carpathy. He says he doesn't even touch the keyboard.

He just uses whisper or some sort of like super whisper app. He said and he just talks to it and just says, "Yeah, adjust the padding. Accept all. " Uh, you know, oh, there's a bug. Fix it.

And uh and and yeah, I mean I I it would be very it'd be super super weird because for the longest time the programmer was, you know, the keyboard was sacred. It was don't have you don't even have a mouse. Everything you do is on the keyboard. You probably have an ergonomic keyboard.

You're dealing with carpal tunnel because you're on the keyboard so much. And if we move to a world where the you still have a genius programmer who can understand the architecture, understand what questions to ask, what features to build, that type of stuff. But the actual interaction day-to-day is is uh is via voice.

That would be a very very big shift. It'd be very interesting. Um anyway, I want to get your reaction to the the the news today that Glean re raised $150 million series F at a $7. 2 billion valuation. Now, will you say Glean? Glean. Yes.

So Glean their uh their whole goal is to uh enterprise search and I'm interested in particularly who are the beneficiaries up and down the stack because I imagine that they're in a position you know to kind of say hey they stuff all your data in glean but they probably want to be more of an integrator uh and sit on top of your data lake on top of your your snowflake or on top of your uh uh you know your AWS installation or your and plug into your Slack.

How are you thinking about um having tools that really allow folks in the enterprise to interface with all of the infrastructure that you have described? Totally.

And you know what's interesting is the the category isn't isn't new like enterprise search and like there were so many startups and there's this graveyard of companies that existed but I think the clear why now or at least the enabling technology was LLM.

So of course glean is going to benefit but like what's powering all of this are open ais and anthropics like really really powerful models. Now there needs to be this layer of infrastructure that exists between the model providers and glean itself.

It's like the hairy problems like memory and caching and so those providers are going to win as well. But ultimately the consumers also and these companies win. It's it's it's amazing even for me like if I'm trying to search something in Google Drive like how long it takes to get something. It's degraded somehow.

It's gotten worse. Like I can't even see my email. It it it's just like there's so many like cookies now that if I search for like Jordy, it'll find that in some random URL string and be giving me some like, you know, spam email I got. It's a mess. Yeah.

But you know, we we should expect and I think uh not I think is OpenAI is going to release a product here. You will have like why why does there need to be an application on top of this? Oh yeah, totally.

I mean, they were they were teasing this a little bit with the uh uh what what was it the um the new the new I think there was a little a little bit of drama around uh different people not wanting other people to invest in Glean OpenAI but then there was also the there was also the the the latest news that um that deep research now can search across GitHub, Google Docs, Gmail, Google Calendar, SharePoint and so that's a competitor.

That's a clean competitor. That's competed. They're just not saying it. And so exactly. And so I think for the applications that make that there is strong really strong product market fit with the large language models.

We should expect the model providers themselves to move up the stack and address parts of these apps because I mean if you think about like the models themselves like yes they make a ton of money but they also burn a lot of cash because OpenAI is just trying to build that next frontier model.

So they're going to go where the consumers are like let me go buy Windsurf right and and be in the IDE. let me go release something in the enterprise search space to compete with vendors like Glean. But so I mean I think it's fascinating. Yeah. Uh I want Yeah. When when Yeah.

When are you ultimately or the CEOs that you're you're meeting with today? When are you know doesn't I think there's areas in which people feel safe, right? If they're working on like a specific application of LLMs with a bunch of deep integrations and workflows, people generally feel safe today.

But is there are you expecting a lot of dialogue around people just trying to assess like am I on the lab's roadmap and and really trying to figure out like at what point they're going to get you know sherlocked. Yeah. You either have to be you have to be building on the right side of AI.

So if you're trying to solve for deficiencies in the model today and you're trying to build these things like very very quickly in a couple of months all of that tooling just didn't need to exist.

So you you just need to be building on the right side of AI as these LLMs become more powerful like your application itself should become a lot more powerful and how do you embed into workflows and also build all of the little hairy infrastructure things to empower all of that.

Yeah, it's interesting like ironically like rappers aggregation aggregating demand being a front door to AI like that's where a lot of the value has accred or at least like it it feels like the foundation model companies feel most threatened by those approaches versus the threat or just want to own them.

they want to own them or they want to compete with them directly as opposed to the okay yes you leapfrogged us on capabilities that's much less of a threat then oh yeah you beat us on a benchmark that matters a lot less than than when people think of this particular AI use case they go to this website instead of mine that's more threatening to the foundation model labs in my opinion totally you know a good analogy is that the hyperscaler hyperscalers like GCP aws and azure are the equivalent of open eye and anthropic for the AI world and and if you saw what AWS did for example right with elastic they open source elastic was a very permissive license and they open sourced it and they started competing with elastic itself what elastic ended up doing was they they went to a much more restrictive license but you still had AWS selling elastic's core product and then Elastic is also existing as whatever it is a $10 billion business today so I Going back to your point, I mean like it's not I don't think it's win or take all.

I I think the model providers are going to have a large portion of revenue and and spend here, but they're also going to be really exciting AI native companies that exist on top. Yeah. Yesterday we talked to Vlad from In Physical working on uh API keys and secret management. Yeah.

Raised a $16 million series A from Elad Gil. I wanted to know more about different strategies and approaches to building uh infrastructure plays that either play directly or alongside open-source strategies. And so uh there's an open source package out there that's gaining traction.

How do you build a product or company around that? Or you have a company that's built something. Do you open source it? What are the different strategies that you're seeing? What's working? What's gone out of fashion? Uh, I just love some color on how to build either with or alongside open source in the enterprise.

Yeah, you know, I I think open source is a double-edged sword.

I mean, obviously we we've been fortunate to back companies like like Hashi Cororp and Click House and we've seen the commercial success of those businesses, but ultimately the for for these companies, the biggest competitor is actually your open- source project, right?

like like why do I need to go pay you millions of dollars when I can just try to do this myself and fork it?

And so for the open source companies out there, I I think you need to be really careful and just kind of understand where your product is going, the types of enterprise features that you could add that that buyers would actually want you to pay for.

Um Aliia data bricks, he has a good analogy here like like building a massive open source business is like hitting two home runs simultaneously. one, you need an a really exciting product where there's a ton of open source traction, and two, you need to figure out monetization.

In the case of data bricks, you know, they had Spark, and it was great, but they had to figure out a way to make Spark really really cheap and easy to use and and and but they had to hit these two massive home runs to build a big business.

So, for the open source founders out there, I would I would say like really consider how you're going to commercialize this product. Yeah. What what are the what are the case studies that people are coming back to today?

I remember learning the the history of Red Hat Linux built on top of Linux of course uh fantastic business wound up a public company uh then we've also heard the story of GitLab that went through Y Combinator became a fantastic business data bricks has a similar story what other stories are people latching on to these days what are kind of like the the the the famous examples that people keep pulling from in this open source enterprise infrastructure world yeah I mean MongoDB is one Look at the success of that business.

And what's really interesting is that they released their cloud product much later from from the actual open source. It actually came right before the IPO and it's actually accelerated their revenue growth since. It's the only company where that's the case in the enterprise software.

So, how were they making money before having a cloud offering? Just Well, they they had an enterprise uh core version of of their open source, but like got it. Then they also released a cloud self-hosted version of that product. Hashi Corp is another example, right, with Vault and all of these systems.

And so there there obviously like companies that have been successful in doing this, but it's you you need to be really careful and you need to think about monetization. Yeah, it just seems so dangerous because you're building on top of an open source product.

You know, this is going to be baked into Amazon and Google and Azure like very quickly, but at the same time, you're maybe a founder mode company.

you have a lot of really aggressive people that can go and move and launch new features that can just keep you a cut above and then yeah uh you know the the company says yeah I could get you know 80% of what I want on AWS but I want it to be perfect and so I'm going with you because you're the best vendor. Yeah.

Basically is is the total cost of ownership of having all of these developers internally and and trying to maintain this in-house is it much less than the actual service that they're providing the enterprise service. If if that's the case, then great. If if not, then you really have to reconsider this the strategy.

Yeah. Nothing's worse than having an open source installation and having to deal with like, oh, we got to turn it off and turn it back on. It's much better to pay someone else to do that. And then and then, you know, it just creates just really interesting dynamics with the open source community, right?

if if I'm only going to release features in the paid enterprise version and I'm going to kind of just stop releasing features to the open source version then your community is going to get upset and so like it's yeah you just have to be careful. Yeah. Well, this has been a fantastic conversation.

Thanks so much for taking the time during the busy day to check in with us. Fun ringing the bell. Come back on again soon. Yes, this is great. Send us some photos. We'd love to see it. Uh enjoy the rest of your day. Big gong. Yeah, big one. We'll see you soon. See you. Bye.

Uh whether you're looking to get exposure to MongoDB or Red Hat Linux, go to public. com investing for those who take it seriously. They got multiasset investing, industryleading yields. They're trusted by millions. Oh, Jordan, you got a new campaign today. How did you sleep last night?

Because I went on a whirlwind tour. I fell asleep in my son's bed. So, I got two hours. It doesn't count. I need Nate's sleep in his bed and then I can have Tyler vibe code up something that merged the two data sets together because I think I got like nine hours yesterday.

I'm feeling fantastic, but I only logged seven hours on my actual eight sleep. So, I got an 86. How'd you do? I got an 83. That's two nights in a row. Oh, two nights in a row. Let's hear it for John Kugan the sleep master. No one outsleeps me. No one outsleeps me. Don't even try.

Oh, we got some breaking news from the printer. Deathly ill. We got some I'm going to be a hand model now. This is uh this is going to be bad. Okay, we got some breaking news. Anderl has been placed number one on the CNBC disruptors list.

Congratulations to Anderl and Trey Stevens is jumping in on the timeline taking a shot at Sam Alman saying, "Ha, take that Sam Alman suck crying emoji. " Of course, they're buddies and and Founders Fund's an investor in OpenAI. So, they're having some fun.

Uh it was crazy that Anderall got number one because OpenAI has what 10 times the market cap at this point. But both fantastic companies. You'll love to see them at the top of the list. Number one and number two.

We should go we we should go through the uh the disruptors list because we have a little bit of time before our next guest. Uh Ki from Linear is joining at 12:45. We have 15 minutes. Let's go through some timeline reaction. Let's go through some CNBC uh disruptors list. I want to see who's going on.

So anyway, uh to to recap the Glean news, they raised $150 million, adding billions to their valuation. They're 70 7. 2 billion valuation now. Uh very interesting to see them 7. 2. And it was a who's who on the cap table. The uh the CEO Arvin Jane really using the long post feature on X.

He gives shout outs to Wellington, Kosla, Bicycle VC, Geod Desert Capital, Archer Man Cap, Altcap, uh, Capital One Ventures, City Ventures, Code 2, General Catalyst, DST, Iconic, Institutional Venture Partners, Kleiner, Latitude Capital, Lightseed, Safoa, everybody's in.

So, this tells me that OpenAI didn't really get their way and not wanting to people to invest.

So what you are referring to is uh the fact that uh back in October of 2024 last year uh there's an article in Reuters that says OpenAI asks investors to avoid five AI startups including Sus uh Ilia Suskgiver's SSI sources say so as as global investors such as Thrive Capital and Tiger Global invest 6.

6 6 billion in open AI. The chat GPT maker sought a commitment beyond just capital. They wanted investors to refrain from funding five companies they perceive as as close competitors. The list includes Anthropic, XAI, Ilaskver's new company,