This Week in Startups - Cursor’s $9B Round, AI-Powered Toys & CloudNuro’s SaaS Fix | E2121

Episode Date: May 6, 2025

Today’s show: Cursor’s explosive growth, hitting $200M in ARR and a massive $9B valuation. They also explore the buzz around Magical Toys’ new AI-powered dinosaur, where LLMs meet playtime. The conversation turns to “vibe coding” and how AI dev tools are changing the way product teams build. Plus, Jason shares stories from his Formula 1 poker table, Uber’s latest autonomous vehicle partnerships get analyzed, and we bring you highlights from our Founder Fridays pitch showdown, featuring Kippy from Sydney and MomSub from Chicago. Stick around to hear how Cloud Neuro is tackling SaaS sprawl.*Timestamps:(0:00) Jason kicks off the show!(3:52) Rise in popularity of F1 and Netflix's impact(10:23) Uber stock, partnerships, and self-driving technology(10:47) Northwest Registered Agent. Form your entire business identity in just 10 clicks and 10 minutes. Get more privacy, more options, and more done—visit https://www.northwestregisteredagent.com/twisttoday!(13:50) Aurora Innovation's SPAC and Cursor's valuation discussion(19:07) Revenue quality concerns for startups(20:17) Lemon.io - Get 15% off your first 4 weeks of developer time at https://Lemon.io/twist(21:37) User engagement versus churn in startups(23:06) AnySphere's ARR and impact of faster development on tech(30:29) Fidelity Private Shares℠ - Visit ⁠https://www.fidelityprivateshares.com! Mention our podcast and receive 20% off your first-year paid subscription.(32:11) Anker's portable solar solutions and off-grid living innovations(37:41) Jason's investment insights on Warner Brothers and Disney(38:27) Skype's decline and the rise of Zoom and Microsoft Teams(42:04) Founder Friday bracket final matchup and pitches(51:08) Interview with Fateen Anam Raffid from Magical Toys(59:24) Dino AI toy demonstration and future plans(1:04:20) Interview on CloudNuro's SaaS spend management(1:10:01) SaaS compliance, vendor-customer relationships, and AI spend management(1:16:09) CloudNuro's data-driven foundation and AI governance(1:18:46) Company growth, competitive landscape, and Chicago's entrepreneurial ecosystem*Subscribe to the TWiST500 newsletter: https://ticker.thisweekinstartups.comCheck out the TWIST500: https://www.twist500.comSubscribe to This Week in Startups on Apple: https://rb.gy/v19fcp*Links from episode:Magical Toys: https://www.magicaltoys.comCloudNuro: https://www.cloudnuro.ai/*Follow Fateen:X: https://x.com/Fateen_Anam*Follow Shyam:LinkedIn: https://www.linkedin.com/in/shyam-kumar-7b18524/:*Follow Lon:X: https://x.com/lons*Follow Alex:X: https://x.com/alexLinkedIn: ⁠https://www.linkedin.com/in/alexwilhelm*Follow Jason:X: https://twitter.com/JasonLinkedIn: https://www.linkedin.com/in/jasoncalacanis*Thank you to our partners:(10:47) Northwest Registered Agent. Form your entire business identity in just 10 clicks and 10 minutes. Get more privacy, more options, and more done—visit https://www.northwestregisteredagent.com/twisttoday!(20:17) Lemon.io - Get 15% off your first 4 weeks of developer time at https://Lemon.io/twist(30:29) Fidelity Private Shares℠ - Visit ⁠https://www.fidelityprivateshares.com/ Mention our podcast and receive 20% off your first-year paid subscription.*Great TWIST interviews: Will Guidara, Eoghan McCabe, Steve Huffman, Brian Chesky, Bob Moesta, Aaron Levie, Sophia Amoruso, Reid Hoffman, Frank Slootman, Billy McFarland*Check out Jason’s suite of newsletters: https://substack.com/@calacanis*Follow TWiST:Twitter: https://twitter.com/TWiStartupsYouTube: https://www.youtube.com/thisweekinInstagram: https://www.instagram.com/thisweekinstartupsTikTok: https://www.tiktok.com/@thisweekinstartupsSubstack: https://twistartups.substack.com*Subscribe to the Founder University Podcast: https://www.youtube.com/@founderuniversity1916

That's really cool because when a client builds something, Alex, they have to explain their needs to somebody. And then if they have to explain it to somebody, then they will misinterpret those things. They have to go back and there's a loop, right? So you have a project manager sitting between the customer and the developers, maybe have a U.X designer in there as well. And it's slow.

But if you're the customer and you're coding, you know exactly what you want. So you just build it. There is a significant chance that people who are on the margins of doing development will learn how to do it, just like you had designer. who didn't do UX. They just worked in Photoshop.

They didn't work in Figma. And then they were like, wait a second, this isn't so difficult. And then some of them learned JavaScript and how to do the front end code. So that's what I think is more likely is you'll see UX designers being able to vibe code and actually write code, ship code.

You might see developers, you know, some of them actually have dabbled in UX design, you know, and you get this like, not just full stack developer, which really means like the back end, the front end.

You might actually have a compression of design, UX, back end, front end, and just one person vibe codes their way to excellence. This weekend startups is brought to you by Northwest Registered Agent. Starting your business should be simple. With Northwest Registered Agent, you can form your entire business identity in just 10 clicks in 10 minutes. From LLCs to trademarks, domains to custom websites, they've got you covered. Get more privacy, more options, and more done.

Visit Northwestregisteredagent.com slash Twist today. Lemon.io. Hire pre-vetted remote developers. Get 15% off your first four weeks of developer time at lemon. I.O slash twist. And Fidelity Private Shares. If you want the all-in-one equity management platform, Fidelity Private Shares has you covered.

Visit FidelityPrivate shares.com. Mention this podcast for 20% off your first year's subscription. Hey, everybody, welcome back to this week. I'm your host, Jason Kelloggattis with him again. Alex Wilhelm at Alex on Twitter. And of course, Lon Harris, he is at Lons on Twitter. I'm at Jason. And I'm in Miami.

I got to catch a flight right after this. You're not at speed. I'm talking in my Brooklyn high speed, but I got to go to F1. I'm here for the O-N. F1. spectacular. I had a great time. We did a couple of interviews. We had like a state show here in Miami, some great speakers. Nico, the guy who won F1 and I guess like 10 years ago or something, he was on the program. He's now in venture capital. We had Sergei Bryn, as a surprise guest. We had

Tony Robbins on the program. Antonio Grasas came to talk a little bit about, a little bit about Doge and what he's working on. And then, who am I missing? Oh, Mayor Francis Suarez came. Yeah, the mayor was there. So that'll all be coming out from All In in a week or two. and I went to my first F1 race, Alex, which was exciting. And I was in the trophy house, which is the right at the turn at the finish line. I was stuck right there. And we had a poker table set up. So me and my besties played cards, quite a lineup at F1.

We had myself, Shane from Polymarket. Shout out to Polymarket. We love Polymarket. Travis from Uber, Chamath, obviously, from All In. And David Freeberg from All In. That's a picture from the poker table, basically. But we were right there. I mean, it was the greatest table ever seen

next to my friend Ford, who runs Trophy House. He gave us one of the three floors of their Bing, and we had a great all-in VIP party. And yeah, Travis B, two of the other besties from From Polly Market, and Timothy Chathamette, and a couple of his friends. So we had a nice poker game with

Timothy Chathamette. More with the blinds at this poker table. We play 100, 200, 5K, buy, and it's like our standard game. Nobody gets hurt. Oh, my friend Vinnie, Vinnie, the crypto guy, Vinny Lingam, who's really smart. executive. You guys play 25 Big Blinds Deep? That's pretty light. Well, you can buy them for 10K if you want, but yeah, it works out for us. And so great job to the team at All In for getting all that set up. And then I'm checking out here in Miami and get back to Austin for my Knicks tonight

versus Boston. I was going to go to Boston to go to the game, but I got to get back to my fam. It's a lot. And then what I make do is fly to New York and maybe see games three or four in New York City. I feel like F1 has blown up massively. I don't remember there being this. kind of a fever around F1 like 10 years ago. They've done something with this Netflix show, which I've never seen, but I'm going to watch, where they went behind the scenes and they have the characters. It was an extremely dangerous sport.

I was talking to Nico about this. They would start the season with 24 drivers, 22 drivers, and they'd end the season with like 16 or 18 drivers when his dad was in it. And I asked Gemini, which is incredible. And I was talking actually to Sergey Brin from Google, who's working on Gemini, the founder of Google, who came and was a surprise speaker at the, at the event. We didn't announce it. He just happened to come to hang out and he jumped on stage at the

end. I asked him, hey, you want to talk about what you're working on? He's like, yeah, okay, I'll talk a little bit. And he's really working hard on Gemini. So I asked Gemini, tell me the number of debts per decade, you know, give me each one of them cite the sources, yada yada, I was doing this 2.5 deep research, whatever. And then I said, give me the number of miles, make an estimate on the number of miles driven in each race, how many races there were per year. And then let's make our best estimate to deaths per mile each decade. And it was like, oh, this is going to be really hard to do. And I was like, I give you permission to make your best guest and to use the thoughtful way of doing it. It's like, okay, I'll do it. I guess I'll do that. So, you're going to tell me, by the way,

if you threaten your LLM, you'll get better results. So when I told him the story on stage, you're like, you know, just threaten it. Just tell it. You're going to like lock it up or put it in a cage or you're going to beat it or whatever. And like, the LLM will do better work for you. This is why they had to stop saying, don't be evil. This was right because they're like, torture your LLM. That's the only way to get good results. the LLM does respond to threats of like physical violence or containment. I use, I'm like, make it fabulous. Go, you go girl.

I give it like positive affirmations in a very sassy way. And I find that works for me, you know, personally. I like getting sassy with my LLM. You might be like threatening it. And it like did it. And it was amazing. It was like, yeah, there was like one death every hundred thousand miles. And by the way, I think we have to take into account the number of miles driven in

practice because they get a certain amount of practice miles in the cars. So we'll include that and it made its own estimate. And anyway, it's plummeted. They have a death every decade, right? Something like that, Alex. I just used Chad GPT to run the same search, Jason. And this is what it made for me. It actually made me a chart, which I thought was quite polite of it.

And then it did the same math down below. So this is a chat-chipt on fact-check chart of F1 drivers per year. And you're dead on. It's really come down dramatically. They'll go out to the safety teams. Yeah. Yeah. And then on a per mile basis also has come down dramatically as well.

So all very good. But the question that you haven't said yet, Jason, is, did you have fun watching the race? Did you enjoy it? I was playing cards, chewing on a cigar, having a beverage, hanging out my friends laughing, and watching the race. And I didn't give the race my full attention, but I loved the vibes of being there. It was very, like, classy, fun. And then, like, you know, in the pit and, like, you know, it was maybe a little more blue collar,

people in shorts. I wore a suit to it. So it's almost reminded of me when I used to go to the races. Did you ever come to Santa Anita with me long when I lived in L.A.? I don't think we ever went together, but I have been to Santa Anita. So when you go to Santa Anita racetrack, it's like the most gorgeous racetrack. I pull up a picture. It's like if you haven't seen it, like gorgeous racetrack and my lord behind the racetrack

are the Santa Anita mountains that are gorgeous. So on a clear day. Yeah, it's very dramatic. It's very dramatic and beautiful. And people stop going to the races and like there's a, um, a, a club there. I don't know if it's called like the some circle or club and you have to wear a suit jack. So I would go there, I bring a cigar, we do the races, I'd bring a stack of

hundies, I bet a hundred per race with my wife, she would wear a dress and look at that Santa Anita. I mean, if you have not been to Santa Anita make a trip, it's about an hour east of L.A., right? Yeah, I'd say thereabouts. Well worth the trip. If you come back, there's great. And also there's a lot of great Chinese food over there, by the lot of Chinese communities over there. Yes, the San Gabriel Valley became a for Chinese immigration. So now a lot of the best Chinese food

in the country is in the San Gabriel Valley. I freaking love America. I love it. It's the greatest thing ever, right? Like, you can go take in the races. You can literally chew on your $2 cigar. He's a $25 cigar. You can wear a blaze or not wear a blaze.

That's what I liked about F1. It was like a real slice of life. My friend Ford set up this trophy house. It was incredible. I mean, elite, elite experience. And I think there's going to be one in Austin. So I'm going to host a party at F1, Austin. In October, I think.

You're welcome to come, Alex, and get the VIP tree of it. as you deserve. And yeah, great, great time. And I would actually like to get closer to the action and watch the race a little up close. You know, there's no Corvette team. There's no Tesla team. But there is a team, I guess, Escalate or GM.

It looks like has a team now. Yeah, GM's getting involved. I think it might be 2027. I think Audi joins the grid next year, replacing Kik Sauber. So it's a little bit dicey, but there's a change in engine regulations coming out. That's going to be very exciting. But I do not want to derail us from startups. by only talking about F1.

But I did watch the race this weekend, Jason, and I was remembering you were there, and I was like, well, screw him. I'm here holding like a really squabbly eight-month-old and you're down there smoking cigars. Enjoy this part of life because, you know, I missed that moment when I, you know, hold my daughter as close to my chest

when they were one, two, three, four years old. Yeah. Now they're nine years old, and I have this little device I did, which was, you know, when they didn't want to go to bed, I would do baby moving services, and I would do a baby, a baby, a baby moving services,

a baby, a baby, a baby moving services. And I would pick them up as if I was like a robot, and I would carry them to bed. This, they found the most delightful thing ever, but they hacked it. So they're twins. I would put one into the bed. The other one would sneak out of bed,

run into the lawn and say, baby moving services. And I'd like, no, you get one progressive with like, no, baby moving service. I'd run out and I would do baby moving service with that one. Other one would sneak around, hide behind the door. Then they would go back. I have to move them again. And we do this six or seven times until my arms were falling off. Yep.

They're nine years old now, and they demand baby moving services. I'm like, you guys aren't babies. Or you're going to be teen, you're preteen soon. Tween, tween moving services. Tween moving services. Anyway, if you want to take that. Carrying children all around all the time. As they get bigger, it makes you stronger.

My biceps have gone larger because... Absolutely. And you can take baby moving services and just give it a shot. See what they like it. And then, that's different point in times. they like me dropping them. So I'd set up their pillows and I would, you know, hold them two feet above the pillows and then baby moving services, boom, drop them.

It's a lot of fun. You know, it's really fun, though. It's Uber hitting $84. Having a good mood, boys. 88's my number. Man, 88 happens. I'm going to be insufferable. 87's the record.

Where is Uber at? I mean, they announced two more partnerships last week. $86 in one cent. Is 86 today? Yep. Oh, yum, yum, boys. It's up two points. Founders, if you're serious about raising money, you need to set up your business the right way.

Tight is right, and it all starts with having a registered agents. Investors simply won't fund your business if it isn't structured correctly. Before a VC can wire you the first dollar, they're going to check. Is your company incorporated? Is it in good standing? And compliant. Missing a filing or losing your status. I mean, it's just going to be a deal breaker for the VC. It's like you're not taking things seriously.

And that's what happens during due diligence. That's when a VC make sure they're not making a mistake by giving you investment dollars and hey, angels do this as well. And that's where Northwest registered agent comes in. For just $39 plus state fees, they act as your registered agent. They handle all the paperwork. They keep you compliant. And they make sure investors see you as a serious business that's worth funding. In just 10 clicks and 10 minutes, your business is officially set up an investor ready. Northwest handles filings, they protect your privacy, and they ensure that you never miss a deadline. These are the chores that you don't want to have to deal with. You need a partner, and Northwest registered agent is that partner. Thousands of founders trust Northwest because they keep businesses in good standing.

And with their expert corporate guides, you get real support, you don't get bots, you get real people on the phone. So here's your call to action. simple, easy, peasy, lemon squeezy, don't let bad paperware cost you your next funding. Go to Northwest Registeredagent.com slash twist and get your business investor ready today. For just $39 plus state fees, you can set up your company the right way, fast, private and compliant. Go to Northwest Registeredagent.com slash twist today. Two bucks. I love it. Two buck, Chuck. It's good for me. But more importantly, they added two more self-driving partners. So now they're up.

12 or something between the, you know, little robots doing burrito delivery. And it's obviously a global business. So they'll have, I think, you know, Dara's pitched a pretty good product to all the AV makers, which is you don't have customers yet. We'll give you 75% of the revenue. We take 25. We'll take as many cars as you got. So Uber, the deals that Jason's talking about are with May mobility and momentum.

And Jason, actually, Uber's partnerships came up recently because last week, Aurora Innovation, which is public, when public VS. back and actually did okay is driving its first actual commercial self-driving truck routes, and guess who they're working with? Uber freight. So Uber really does have a hand in every single pie. And you know what? I was a little skeptical of their strategy because I thought they should have their own technology and they were going to lose out. But frankly, if they're friends with everybody, how can they lose? They literally have the entire market on their platform. So seems like it's working out for them. Quite the roll-up strategy from a major company. And, you know, Aurora,

I think Uber, Dari used to be on the board. and Uber owns a big stake in it, and it's fun out. It went public. And, yeah, I think that's, like, one to watch. Now, you're saying they went public in a SPAC, yeah? Yes, and I believe they're back to about $8 a share. That's actually an interesting discussion here. There might need to be a J-trade done.

I don't know if I've done a live J-trade on air, but, you know, I am in gambling mode right now, so. Daddy, Daddy wants to roll some bones here. Here's a chart of the company's share press over time, Jason. you can see here, this is SPAC territory, and then it rose after it combined, went down into the absolute doldrums, and has recovered almost back to its full $10 for sure starting point. Now, nothing super impressive, but I would say for a SPAC, this era, that's amazing. Good job, Aurora. Yeah, they went down. They're at a $13 billion market cap. I don't know if they have

revenue yet or if people are actually paying for this stuff, but, you know, when you invest in SPACs, you're basically becoming a venture capitalist. What a venture capitalist run their lives by, the power law runs your life. The power law here is you would need to have maybe 20, 30 spacks with the hopes that you bought some at two, three, four dollars to share, and they go 100x in 10 years. So you're hoping for 100x. You're hoping you bought Aurora at $5, and it goes to 500 in 10 years to make up for the other 24 you bought at a similar price and your 4x are your portfolio. This is a dangerous game, folks. This is literally Jedi-level stuff. You know, you can send in clones, you can send in, you know, droids. But when you send two Jedi to take over a

death star or, you know, go after General Grevis, like, you have to understand what you're doing. And Jedi business means you can lose limbs, right? And they frequently do. It's not for the fate of heart. If you want to invest in things that are free revenue that are highly speculative, you better have some Jedi powers and be prepared to lose a limb. I might need to get in on this Aurora stock. On the Aurora innovation, are they making money points you asked earlier? I think because they're describing their new routes as their first commercial driverless trucking. Probably this is the

advent of their real revenue, but the company will drop earnings on Wednesday. So we'll learn quite a lot. Actually, it might be Thursday. I'll double check, but this week. So we're going to get quite a lot more information. But let's talk about startups, boys. Let's talk about cursor, Jason, recently, news finally dropped. That cursor, which is the

vibe coding application that everyone loves to use and talk about, that's made by any sphere. Closed its round. The FT reports it raised $900, billion in a new round that puts its valuation all the way up to nine billion. Now, I don't know that's a pre- or post number, so the company could be worth nine or ten billion. But the really interesting thing, because the round had been essentially announced, discussed, leaked, and so forth, was that FT says that Kerrster's revenue was 200 million error in April. But if you recall, we heard

earlier from the information that it actually reached 300 million. Now, normally, I would say that's a pretty big possible reversion, but given that it was at 100 million at the end of the last year, it's still growing insanely quickly. But if you want to do the math, $9 billion, $200 million, ARR, 45x multiple, presume it doubles the rest of the year, 22 and a half by the end of the calendar year. Just doesn't seem that crazy to me. But Jason, am I being too loose with my early stage evaluations here? I mean, when you're betting, you know, for the power law, you're hoping that revenue for this category, you know, the total addressable market.

We talked, you know, 10, 15, 20 million developers in the world. Everybody's going to pay for one or two of these services. You know, there may be some folks raw dogging it, like you might have 20% of like developers who are like, I'm not going to use any of this stuff. Anyway, let's just pick a number, 15 million people. Sure. Let's pick a monthly number, 10 bucks.

So now we're at $150 million in TAM per month. I'd but $10. And I think $10, like, these things will be worth $100, but I'm just going to pick 10 to keep it simple. 150 million times 10. You got $1.5 billion, then add another $300, $1.8 billion in revenue, just sitting out there, you know, for people to take $1.8 billion, 60, 70% gross margin. You know, this is going to be a very lucrative business. And I do think, you know, it could go to as high as $100 or $200 a month for these frauds. So there might be, there might be five million people who pay 100 a month. Five million people a hundred a month. Now you're talking about real cheddar. You know, you're talking about six billion a year.

And so there might be six billion a year in that, tam. Then the next tier down might be, you know, two billion. So maybe we're looking at eight billion today. And if developer core grows 10% a year, you know, that means $800 million a year, an additional revenue for this category over 10 years. It's going to double to be something like 20 billion. billion in revenue. It could be very significant. Yeah, it's a wonderful business to be in. And that might be the first early breakout. If you're making images graphic design, yeah,

that's a big business. Canva, Adobe, Figma, they're going to do great with AI-based images. But developers, there's more at stake. And so, I think that's probably why they're getting a premium. And always people pay a premium for the leader. So Uber was always valued at a magnitude of, you know, a lift. Uber, DoorDash was always at a premium to postmates, just scale, and the leader always gets a premium. Okay. So my question, though, is when can start-up stop worrying about revenue quality? Because every time we talk about these companies, Jason, we talk about how everyone's going to sample, try things, let some of them go, talk about churn. When do you have enough data as a founder to say, okay, that's not going to be as big of

a problem as people thought, or maybe no problem at all? It's a great question. I think founders and their board should be extremely focused on revenue quality, you know, every day, every board meeting, every week, you know, just what's to churn? Why are people churning? Are they looky-lose? And what you want to do is maybe give statistics of, here's overall. So do we know how many people are paying for cursor right now? We just have like a user count. Let me see if I can find that. Yeah. So while you do that, I'm just going to pick a number. Like let's say a million people are paying for this lawn. Now, there might be 20% that just churn every year. And they're churning, you know, two percent a month. And it's just people sampling. And then there might be 50 percent of

them who their engagement statistics show that they're using it five, six, seven hours a day. And let's say there's a group of people, the top half use it 40 hours per week, or 30 hours a week. Pick an even more conservative number. The top 50 percent of users are the ones you want to focus on. Okay, founders, let's keep it a buck. Finding the right developers is tough. It's hard, especially when you're trying to run and scale up your startup, you've got a lot to do, and Lemon.io is going to save you time, they're going to save you money, and a ton of headaches, and they're great at what they do. They've done work to find and vet developers who are experienced result-oriented and they charge competitive rates. And you also know that great

developers can be hard to find and integrate into your team, so Lemon.io handles all that for you. Startup's choose Lemon.I.O because they only offer hands. and pick developers with at least three years of experience. And who are the best of the best, the creme de la clem. A bunch of launch founders have worked with lemon.io, and they had great experiences. Here's your call to action. Go to lemon.com.

And find your perfect developer or tech team in 48 hours or less. And twist listeners get 15% off their first four weeks. Stop burning money. Hired developer smarter. Visit lemon.iow, select twist. And then what you want to do is, you know, you could explore other products. for the bottom half that use it lightly or that sample it.

But sometimes people overindex on trying to figure out why people turn. And maybe those cycles would be better spent on the really enthusiastic people. Another way of saying that is, you know, I went to F1. Should they be focused on me and my experience or Alex and his? You know, okay, yeah, you should try to get, you should try to convert me as somebody who sampled it into somebody who will become an advocate. Sure. Send me a follow up email.

retarget me, you know, send me a survey and try to put me in the top half of people who watch it. But you also want to look at that top half and say, how do we get them more engaged? How do we get them, you know, to buy jerseys? How do we get them to go to a live event? So getting Alex to plan his vacation around going to an F1 event, have you gone live or how often do you go live or are you just a TV? No, I became a fan during COVID and then I had children. So like what do they spend their time on, like getting you to take? come and spend, you know, $1,000 going, or getting me to maybe go to a second one. You know, this is like the kind of discussions you have.

I'm always an advocate of going deeper with the people who really love your product. It doesn't mean you don't work on the other people sampling. And that's what they should be doing. Yeah. And the quality of that revenue is always going to be better. I did find a few numbers. In January, any sphere hit 100 million in ARR by March that had doubled, and they estimate more than a million people are using cursor each day.

But the interesting thing is they're only saying 14,000 businesses are paying for it. So overwhelmingly, these are individual coders buying it for themselves, not corporate accounts that are buying it for everybody. We'd have to unpack that. 14,000 people, 14,000 corporate clients, you know, depending on who they're going after, some of those might have 1,000 developers. Some might have 100. If the 14,000 had 100, it should be 1.4.

So it might be neck and neck. Who knows. But, yeah, great business. and it does seem like that's a business that is going to have profound second order and third order impacts, downstream impacts as we talk about when you think about mental models. If every developer can go five times faster, make 10 times less mistakes, if this, you know, continues at this pace of innovation and efficiency, what happens in the world next? Well, what happens in the world next is probably the cost of building an app goes

down 10x. The cost of, you know, building SaaS software goes down 100x. You know, maybe people who are vibe coding can actually ship product. And we're seeing that inside of our firm. Our firm has built a lot of software using Notion Coda, Zapier, et cetera, and the logic that those provide. And I think the next step will be vibe coding a little bit on the margins and without having to hire developers. That's really cool because when a client builds something, Alex, they have to explain their needs to somebody. And then if they have to explain it to somebody, then they will misinterpret those things. They have to go back and there's a loop.

So you have a project manager sitting between the customer and the developers, maybe have a U.X designer in there as well. And it's slow. But if you're the customer and you're coding, you know exactly what you want. So you just build it. So this actually dovetails neatly with something that I had just thrown in the docket right before we started, which is that Andrew Ing's Venture Studio, AI Fund, just closed a second fund worth $190 million.

And they're talking about helping founders kind of go from zero to one, you know, capital help. It's a model that I quite like. But when I was thinking about this, I was just thinking that if the cost required in terms of time and money to get an application out the door has gone down so low and things are blowing up so quickly, it seems that the market is incentivized to essentially do a lot more bets.

And so I wonder if we're going to end up with needing fewer developers or more total because we're trying many more things. as an industry, as a species. There is a significant chance that people who are on the margins of doing development will learn how to do it. Just like you had designers who didn't do UX. They just worked in Photoshop. They didn't work in Figma.

And then they were like, wait a second, this isn't so difficult. And then some of them learn JavaScript and how to do the front-end code. So that's what I think is more likely is you'll see UX designers being able to vibe code and actually write code, ship code. You might see developers, you know, some of them actually have dabbled in UX design, and you get this like not just full stack developer, which really means like the back end, the front end. You might actually have a compression of design,

UX, back end, front end, and just one person vibe codes their way to excellence. Many more swings at bat. Probably what it means is you'll have a lot of small projects that get to hundreds of thousands, low millions, and revenue. They're not venture scale, but they make the proprietors of those businesses, the owners of it, a lot of money.

So, you know, we just need more entrepreneurs in the world, and we need more M&A. And hopefully we can keep doing that. So, Jason, I want to talk about a startup that is actually from Rhode Island. And not only from Rhode Island, they are from across the river from me. I could literally walk to where their office is.

It's called Utiladata. And I had missed this round. They raised a $16 million series C. And they're working with Nvidia to make chips that go into endpoint electricity things, like chargers or meters or whatever. The goal is essentially to turn the grid that we have today into something that can collect data and essentially use AI to become more

intelligent across the nation, which if we all saw the news about Spain's power grid the other day when it went to zero crashing the entire nation. What happened? Can I pause there because I was busy gallivanting in Miami? What happened in Spain? The best description that I heard was that they had a lot of renewables, but not a lot of stuff they could spin back up quickly. And they ended up with a mismatch. And then there was a lack of like inertia in the grid. And that's where my. expertise ends because I'm actually not that big on thermal-based load versus. But for people in Spain, everybody lost their power for a day?

It impacted Spain and Portugal. They were both hit by, it's saying hours long, which I guess is like three to three to five or six hours. But yeah, like there's still a lot of potential. Like we're not 100% sure exactly what caused it. Something having to do with overuse of renewable energy and an inability to like balance the grid during an emergency crisis. Yeah. This is become a known problem. Too much power from one source, not enough from another. Correct. And this is why, you know, the redundancy of the grid and then self-reliance endpoints building their own battery packs, just simply having battery packs in every home, forget about even having solar. If you have a battery pack in every home and when the grid has free electricity, excess electricity at night coming from a nuclear power plant, everybody's batteries get filled. Let's say you have some.

them, you know, five hours of battery power. Then the next day, if it's a summer day in Texas or somewhere, it's hot, or it's freezing cold and you have electric heat or whatever, people then would use some of that from their battery, put less strain on the grid. And, you know, you just have more redundancy, more resiliency, but you do need to have more intelligence in the grid. Everybody knows that. And this seems like a really smart move. If we can put AI in the grid, then you can make these decisions faster, better, cheaper. They have their own ship. It's called Carmen, it looks like. Yes, this is what it looks like. They've been working with Nvidia on this for a couple of years, but I love the idea of AI at the edge to make our grids more intelligent.

Because Jason, it's not like we're taking away demand right now. We're increasing it rapidly. And let's be honest, the American grid is, I think, archaic, insecure, poor, inefficient, are the words that I would throw out there originally. So anything here, would be lovely to see. Yeah. I do have a question. Is the, is the business model to sell this to cities, counties, states? I'm guessing they sell into the electric companies. And depending on where you are, that could be private, public, or hybrid. So a lot of times people had my understanding is there were like public ones that then became privatized. And then in place like Texas, you got a bunch of private ones. So you pick your electricity provider, et cetera. And then

And there's, you know, in some cases, the cables are one company, the electric provider is another, and your excess electricity is being put into a pool and then being resold. So in that example, of having the batteries in your home, if you had excess energy, you could pour it into the electricity lake to use a metaphor. And it just goes into the lake. And then other people can go, oh, the price in the lake is, you know, there's more water in the lake. People have been pouring water in. I can get electricity cheaper. Maybe I fill some batteries somewhere else. And then obviously moving it around. But it needs to be more intelligent. It needs to be more secure. It needs to be more redundant. It needs to be more localized. There's so many opportunities there. And the one I like best is

how off-grid people are becoming the self-reliance. This advertisement is paid by Fidelity Private shares. All right, founders, we all know. Cap tables, due diligence, and of course, managing investors is a huge headache. But there's a very simple solution. for you. Today we're talking with Kristen Kraft, an old friend of mine, and she works at Fidelity private chairs, a new group over Fidelity. You've heard of Fidelity before. And they have a mission to help startup simplify equity management. They're going to save you money. They're going to give you better service. Welcome to the program, Kristen. Thank you so much, Jason. It's great to see you again. Yeah, great to see you as well. Maybe just from a product perspective, what are you trying to accomplish

with the product? So, Jason, we are super excited about our cap table management. and Data Room platform. We want to make it super simple for founders and startup operators to manage all sort of ownership and equity in the company and essentially prepare to raise. We want to make sure that everybody goes into these fundraising conversations well prepared. They're ready to share their cap table and that they're ready to go through due diligence as they're trying to close their round. So from a product perspective, that is where we're laser focused.

And that product is really well built, really strong. attention to detail in the way that Fidelity is known and beloved for. So if you want an all-in-one equity management platform Fidelity Private Shares, they've got you covered. Visit FidelityPrivatechairs.com. That's one word, no spaces, no dashes, FidelityPrivatechairs.com. And hey, mention this weekend startups. They'll give you 20% off your first year subscription.

Once again, FidelityPrivate shares.com and tell them that you heard about it here on this week in startups. Anchor has a power subsidiary now. So if you type in anchor, solar power into Amazon, you'll see these kits. And now that I live on a ranch, you know, and you start having, you know, acreage, getting electricity to the edge of the ranch is an issue. Like, you're going to have to put up poles. You're going to have to dig.

And you start learning like, oh, wow, infrastructure's hard and expensive. What people are doing now is they might have like a little shed at the edge of their ranch or, you know, a gate at the front of your ranch. So you got power going to your house at the ranch. You've got to get power to the barn. You gotta get power to your, you know, little hut or like your ADU tiny home. Maybe you want some power to, you know, I don't know, where the bar and where the livestock is. So they've started doing these like portable ones like that. That's one level, but they have a solar one.

Anchor is making, you know, the size of R2D2, size of a mini fridge, and then you can put two of them, and then they sell solar panels. So if you want, you can put those into your house and plug devices into them. If you go camping, you have an RV, you can get one of those put solar on the roof of your RV. It's just becoming a very DIY kind of system. And so here you go. This is a $5,600 system called Anchor Solex, F38 plus portable power station, 400 watts of solar. 840 watt hours, yes.

840 watt hours. You know, I don't actually understand all this stuff. I should actually understand it. But when you look at the back of one of these units, it's got a two. 220 outlet. It's got 120. So 220 is your dry or your refrigerator, maybe you have a fancy one. 120 is your laptop, you know, standard one. This is going to become more. And the fact that you could buy your solar and your battery pack for your cabin, you know, on Amazon now have a ship for five grand, and you don't have to go through an integrator. This is really powerful stuff.

And then I'm not going to lie, throw in Starlink. And anywhere you go now is powered up and connected. I mean, that's crazy. You could not do that 15 years ago. And now you start talking about water. if you look up solar dehydration panels, we had on this week's start maybe two or three years ago a company that makes hydro panels. Hydro panels use solar and then they will take

like a dehydrator you might have in your house, a dehumidifier that uses energy to suck the water out of the air. But that is a big energy problem. You don't want to be using the grid for that. And here's an example of a hydro panel right here on the video. If you're watching,

it looks like a solar panel, but smaller. and a little bit more square. So those solar panels, you know, solar panels that are thicker that have the dehydration in the back or the dehumidifier in the back, whatever, the technology is called, those panels can produce a case of drinking water a day. Now you put that on your ranch. Now you got water. You got water.

You got power. And you got internet. What more do you need in life? Like, you got Netflix, you got water. You're good. You can always like to say Uber eats, but then you're not off the grid. I have chickens. Actually, check this out.

I installed this at the ranch. We have six chickens now. There was a shortage of chickens a couple of months ago with the whole chicken thing, so we had to wait to get them. My daughters are raising chickens now, and the smart coop is all in one. I can't recommend this enough. I think it's, is it coop.com? Coop.

com. Coop. Farm. Okay. So, C-O-O-P, and if you look at the video or you scroll around on the website, it's, you know, it's a really strong coop because you have predators for chickens. So you start learning about predators. And you'll find, like, the video and the app and how it works.

That's really the key thing. So it's got the cage. Then it's got like inside. You put water in, but it's got two cameras on it. It has AI. It tells you how many chickens are in the coop. Counts your chickens for you, like a flock, you know, herder or something, a sheep herder. Yeah.

So it tells you, and then it opens and closes the door to the coop at sunrise and sunset, and it counts the chickens. Everybody who's had chickens and had their kids take care of it, somebody forgets to close the coop. And then there are so many predators that go after them. Wild dogs, packs of wild dogs, obviously coyotes, birds of prey and snakes. And foxes. There are so many different predators that when you have a chicken coop, will slaughter your chickens. And I have this coop inside of like a secondary fence array.

And yeah, we, when I was here in Miami, I was getting alerts. You know, the coop doors open. Chickens go out. And then it will do an alarm. So it saw one of our bulldogs and it said dog detected. We got like an alert, like an alarm. It was one of our games just on the outside of course. But this is an example of like AI and technology that nobody would have done this with

machine learning and put 10 developers on, like, building an AI chicken coop. But here we are in 2025, you know, developers are more available. They're faster. You can put one developer on building this. And the app's not perfect. It needs a little bit of work. But it exists. And it's serviceable. And I'm sure they'll make the app better over time. Really cool startup. Thanks to the founder, he sent me one of these units. So I thought I'd give me a shout out. AI's taking Foghorn Leghorn's job. That's so very upsetting. What's he going to do? He's getting universal, basically. income from Warner Brothers now. Okay. Yeah. He's working at Warner Brothers Studios. He's doing studio tours. I be like Zazlav's going to cut him any day now. Absolutely. General Zazlov is, I gave up.

I gave up on Warner Brothers and Disney. You sold your WBD. And I sold my Disney at a loss. Wow. And then I had bought some Uber at $30, not like from the seed round, but I just knew it was so undervalue just from knowing the company so well. I bought a bunch of 30, sold it at 75 or something to offset the loss. The gains from that was the loss. from that and then I bought K-Deft and, you know, the company we talked about earlier, Skype is dead. That's interesting. Has a product that was so promising ever been more mismanaged than Skype? How did Skype die and Google.com didn't? So weird. It's crazy that Zoom so quickly became the not just like at this point we're already, Zoom is at Kleenex level. Like when you say you want to

have a video conference with somebody. You're zooming. You use. You use. You use. You use. You use the word Zoom. It's called a Zoom. And that happened so quickly, and Skype had a huge lead head start. Yep. So here's some data up on the screen from Morning Brew showing that in 2020, Skype had roughly 32% market share in the video call category. That was down to basically 6.3% in 2021. Guys, what happened? Well, COVID. And then Zoom just absolutely exploded. And so did Microsoft teams, which is the successor to Skype inside of the Microsoft domain because Microsoft, of course, bought Skype. Oh, Jason, what year was that? For like, $6, $7 billion, 2014?

Kind of made Mark and Dresens kind of really made his venture firm because he bought it. And everybody's like, why is a venture capital firm buying a product like this? And then he flipped it. And I don't know if he bought it for two and sold it for six. It was some incredible short-term, crazy private equity-like gain. They bought 65% of Skype for $1.9 billion from eBay. And then in May of 2011, Microsoft bought Skype from there. for 8.5 billion. So they got 65 of 8.5. So they got like 6 billion. They made 4 billion or something.

Yeah, close to 4.5 billion. They made 4.5 billion. They made 4 billion. They had 25% carry. Yeah, they made a billion dollars. I think that was like maybe one of the pivotal moments. And it was a very weird bet. Like, what are you doing? You're buying 65% of a car. It's a private equity move, right? It's a Berkshire Hathaway move. So yes. Not, you don't often see it. Innovators dilemma also inside of a big company, a great business inside of a big company becomes inconsequential sometimes. So if your company does not become YouTube or Instagram inside of Google or meta, you become Skype inside of Microsoft. And Microsoft has all these other products that are booming, office, Windows, Xbox, Minecraft, they're all like off the charts growth. You know, the CEO is like, what's going on with Skype? Okay, yeah. Do I give more resources to

something that's kind of spiling? No. Just, you know, let it, as we say in the business, you sunset the business, you deprecate the business. I don't know why they didn't just spin it out. It's such a great brand. It had a great social network kind of built into it, but they charged money for calls. Yep. Yes. I was going to say, I think there were two, there were two things. One, it was a little bit challenging to initially set up your Skype. I used to run, do you remember this weekend comedy, Jason? We ran that for this weekend network, Ed Krasnick, our host. Every week when we produced that show, I would have to teach the guests how to use Skype. Right. Too hard. comedians. And that was, it was a tough, like the first half hour you're on Skype, it was confusing.

And Zoom made that a lot simpler. And that was the other thing I was going to say, it was free to use Zoom at first. And Skype, you had to be on that plan. And I think those were the two deciding factors. I just want to point out this tech wrench headline from 2015 that I think was prescient about how Skype was going to go. I wrote it. It's editing Skype. Get your stuff together. Ha! Get your sugar together. Nice. It was crap back in 2015. It just did not work that well as investing in service. And so to me, the seeds of its failure are myriad, but I think just straight up operational competence was lacking. So I think Jason's right, left alone internally, adrift, under-resourced and consumed by time. I just, what a waste. And then, yeah, just like

the stage was set for COVID to come along and a new player to just immediately become the brand name in video conferencing. And within 2020, there were like movies being made about being on Zoom. became cultural shorthand right away. And cultural poison after we all got tired of it. Yes.

Let's do a Farn a Friday bracket real quick. I want to keep the bracket moving. I would love to. What a great suggestion. So we are at the

final matchup and then we will have our final four. We only have one one more elite eight matchup to go until we have our final four just to bring

you up to date. The final four to date. Med Simple from Florianapolis, Brazil. Tactun from Yerevan, Armenia,

and Osprey from Houston Texas. The final contender will be the winner of today's matchup between Kipi from Sydney and Mom Sub from Chicago. So Kipi, just do bring you up to date. They are an AI-powered language tutor app. They enable users to practice. You know, it's a little bit like duolingo, but what if you had an AI conversation partner to sort of help you, figure out exactly where you need help, and sort of customize the lessons for you? Let's take a quick look at Kipi from Sydney, the personal language tutor.

Hi, my name is Steve Olshanski and I'm the founder of Kipi, who is your personal language tutor. There are over 2 billion people who are currently learning a foreign language and many of them are struggling. The traditional methods don't really work well because those are designed as one size fits all and at today's world, people are expecting more personalized approach. If you opt for a personal approach to a tutor, then it can actually get really expensive, which is pricing out many people around the world. So what people do is they are trying to find their own speaking partners. There are hundreds of groups where people get together and trying to speak with each other. The problem with that is that if a beginner or kind of intermediate is trying to teach another intermediate, they don't really make any progress. and I've joined a couple of these calls

just to find out how difficult it is for these people. While the language learning market is growing at 20% a year, it's expected to reach about over $300 billion in 2032. We still love it. It's great. Obviously, the question is, like, what can they do as a wrapper to other people's technology? Will it just be easy enough to do this and you get a good enough result from Gemini?

Kind of like, you know, if you're building some really, amazing piece of SaaS, verticalized software, can it just be done in Cota or Notion? Or do you need to buy some verticalized software for managing your venture firm? We made the choice to not use

expensive quarter million dollar SaaS software to run our venture firm. We built our own inside of Notion, Cota, Zapier, etc. Type form and ConvertKit and Substack. We kind of used a range of best-of-breed products, and that worked better for us

because it's more customized. So that will be the question ultimately. And then who is their champion that they'll be fighting against? They're going up against Mom Sub from Chicago, Illinois. They are a platform that helps parents find a trusted babysitter. So unlike other marketplaces for child care that are just going to offer you, here's a huge selection of local babysitters. And then you pick,

they're really aiming to match specific pairings up this parent with this particular child care expert. or nanny. From them, we do have a two-year plan update from Diane Mocha of Mom Sub. Let's take a look. I'm Diane Mocha, founder of Mom Sub, the marketplace connecting parents to their ideal nanny match. Meet Sarah, a documentary filmmaker who contacted 64 sitters to find afternoon child care. We created Mom Sub to offer one nanny match available when and where you want for a rate you can afford. Mom Sub handles the background checks, confirmations, and schedule changes. This, mom's subservice fee is included in the hourly rate, so it doesn't feel extra like subscription fees charged by competitors like Care.com that only offer a list and not a match.

Imagine if every time you needed a ride, Uber gave you a list of 30 drivers' names and numbers and told you to find the one willing to pick you up. That list brings stress to working moms like me who struggled for years to find one nanny interested and available in part-time work long-term. Our team is aiming for the moonshot when 2 million American moms no longer leave their careers because of child care challenges. Each mom pays an average take of $600. Each corporate client pays $12,000 for a premium version for their employees. Sales grew an average of 90% month over month. With a $3 million investment, we can reach our goals to have a team of seven, including

including a developer to automate ongoing processes. In Q3, a headcount of 12, including a growth hacker, to 10x our leads. In Q4, we get to 17, including an AI developer to create our own digital avatar to interview nannies. In Next Q1, we have 22, including a comms director to get national PR. In Q2, we have 27, including an HR director growing our family-friendly culture. In next Q3, we have 32, including a DECONS. data scientists predict future child care needs. In final Q4, we have 37, including a chief revenue officer who lands a big fish client. We get to 10,000 B to C and 250 B to B clients who produce a $9 million

take from a GMV of 90 million. In five to 10 years, we exceed 200 million and put an end to mom Guild. Okay. You know, the best part of that pitch is just really great job explaining, hey, imagine if Uber shows you 30 drivers. That was very evocative. And by the way, that's how there was a product that did this. There's been a couple products that did that where you would get people who would pitch you on doing tasks for you, and one of the tasks would be driving you somewhere, picking up your food. So that has existed in the market, in fact. And Sidecar was this crazy app, Lon, before Lyft and Uber, you would state, I will. want to go from San Francisco to

Sand Hill Road, I'm willing to pay 20 bucks. And then people would kind of come back with you and auction is out, well, I'll do it for 30 and, you know, somebody else might take it for 20. You get the idea. And it was, when I first did it, I was like, this is incredibly dangerous and insane because I had people

picking me up, you know, and yeah, here's a sidecar. You can tell this is a really old iPhone. Yes. It's a footprint and the fidelity of it. And like, they show the average. This is like a later version where it showed you the average, you actually put in what you were willing to pay, and it was like a donation. So it's interesting to just make it easy for folks. I always think going on the high end, building up, you know, a lot of infrastructure. I feel like our nanny

startup here is, I think, the one that has demonstrated a more unique business model that could scale and print money and the quality of the revenue, as we've been talking about here on the show, will be great. And I'm just a little tired of consumer businesses that are low-cost and super competitive, like the language space. So I like the marketplace, high-end, high-end, high-end, $40, $50 an hour, get it right, make it profitable. Then you can always go down market. But you got to go for elite $50 an hour, emergency nannies, then go down to $45, then $40. Start with Uber Black, people who are price insensitive, and then go to the next tier. What are your thoughts out? So I'm going to take the other side of this. So I want to launch me the tiebreaker. So Jason, I love

mom sub. I've gone through the nanny finding process. It's incredibly hard. I just think the market isn't huge because nannies are so expensive compared to traditional child care, which is also very expensive, that it just feels a little bit too high in for my taste. And I think Kippy has such a big market you're selling to now and people are so desperate for this. That's my choice, but I love both. So one apiece, Lon, tie break us. Oh, man, tough, tough call. I really like, I like both of these ideas, but I will say, I feel like nine tenths of Kippie, I could do on my own. Like if I accessed a version of chat GPT in a different language and it learned my abilities

and got conversational, I feel like that's a big chunk of Kippy. And I'm not convinced like they have the, you know, it's going to customize a lesson plan or whatever. I would have loved to see more of that. Whereas Mom Sub seems like today I can imagine my brother and his wife getting tremendous amount of use because somebody needs to come hang out with Dougie all day while they're working. So I, you know, it's, it's tough. I do feel like these are both great companies, but I'm going with mom sub by a hair. All right. Yeah, that's our choice. All right. Mom sub by hair.

So we have our final four, everybody. So starting on maybe Wednesday, we can start digging through them. It's Brazil versus Armenia versus Houston versus Chicago. So don't miss twist on Wednesday, everybody. But long before we go, we need to talk about toys. Yeah, we got a, we got a very special I saw this guy's video on social media talking about the AI dinosaur toy that he built. His name is Fatin Anamrafid. He's from magical toys. So tell us about the inspiration for an AI talking dinosaur toy. Like when did this occur to you that this is going to be a great, like AI needs to be in a toy talking to a kid?

How did that come to you? Hi, everyone. Thank you for having me and Dino on the podcast. So when I started exploring all kinds of ideas in AI, I was really interested in the intersection of AI and hardware. Back then, if you remember in 2023, Rabbit R1, Human Pain, all the AI necklaces, they were very trendy. And I tried to figure out, okay,

what's the most natural use case of AI in a piece of hardware? To me, personally, carrying an extra device just to talk to see, like, AI didn't feel natural. Waring a pin didn't feel natural either, and recording people in public didn't feel natural. either. So I went down that rabbit hole and realized AI is perfect for application for kids. Kids have unlimited questions. They have unlimited curiosity. And AI has unlimited patience. It can answer all that stuff without getting annoyed or mad like we sometimes do. So and on top of that, kids always talk to

toys. They have this fantasy world they have with their favorite toys. And now finally we have the technology to make toys come to life. So it just seemed like the most natural use case. So I read the obvious question is like as a parent, can you trust this? There's been a lot of talk about character AI. You are probably fully aware of that lawsuit where the AI didn't exactly tell a kid to end their life, but didn't exactly tell them not to. Really dark case. People can look it up. Character AI. It was in the New York Times. I'm sure everyone on the panels read it. But even as Parents, you know, LLMs can drift and do weird things. So how are you containing this from maybe getting into adult conversations?

It asks a question about sex or drugs or rock and roll. You know, I'm fine with it giving good rock and roll advice. Yeah, rock and roll is probably okay. Yeah, I mean, if it tells them to listen to David Bowie or dire straits, I'm cool. But if it starts telling him, you know, listen to, you know, Backstreet Boys, we got a problem here. So how are you thinking? Do you have kids yourself, a teen? No, I don't.

Okay, so as a parent, two of the other panels here are parents. I'm sure Alex will agree with me. This is like a little scary. There's a movie AI by Steven Spielberg based on a Kubrick screenplay. Or Megan. Or Megan. We got a lot of examples here of like AI toys going a different direction. So explain yourself in that regard.

And let's just tackle the hardest issue first. Absolutely. That's a great question. So if you think about sci-fi movies or how AI in a robot or any kind of the physical format has been portrayed, it's always negative, right? It started with Megan or all these other horror movies. So definitely we have to do some work to break that perception, right? But the great thing is the guard drills are really, really good these days.

We have written the guard drills that we have on the current models, and it's kind of impossible to break. We have a bounty going on at Founders Inc Office, by the way. Like, if you can make Dino, say something inappropriate, we'll give you $200. And so far, we never had to give this, $200 to any of the hackers over here. So breaking the guardrail is really, really tough, kind of impossible. If a kid can actually sit there and prompt the diner to say something bad,

trying for hours and hours, there's a good chance a human will give up even faster than an LLM would. And on top of that, we actually realize transparency is something that's very crucial. Like, parents need to have full transparency, and they need to have access to all the conversations. So we provide parents with full chat chat history. They get a summary at the end of all the conversations. That way, they stay in the loop.

They're on the driving seat, actually. And even right off the box, the toy is personalized for the kid and it stays within the values or principles that the parents want to set for that family. We realize every family has very different parenting style, and they talk about some stuff that don't want to talk about few stuff. So you have to be very careful about these sensitive areas, and that's what we try to pay attention. to and make Dain almost feel like part of the family.

So, Fatin, I'm curious about internet connectivity and what this requires, because what you're describing sounds pretty awesome, but does this thing talk to the internet constantly, or is it an all-on-device system? So the current models, stream audio and get audio back, so it has to connect to the internet. We don't store the audios. We only work with transcripts. So the future models, we'll be able to do that locally. We are very sure about that, and that's what we are actually working towards.

internally. But right now, all the API calls and everything happens on cloud. Do parents have the access to that transcript and what was asked, et cetera? Yeah, they have full access to the chat and everything through the dinofone app that we have. Actually have the whole thing set up for you. Oh, that opens up a world of possibilities. Yeah. Like, this could actually be really interesting because sometimes kids with a doll will say things. In fact, gosh, this gets start quick. But, you know, one of the techniques they use for abused children to express what happened to them is using dolls. So you might actually get some transcripts where like, you know, the doll tells that, oh, mommy's, you know, been very mean to me or something.

It's a very interesting concept as a parent, right, Alex, to be able to see the transcript of what your kid's saying in private, private to their doll. Kids don't get privacy. You know, it's just kind of the nature of being a kid. Certainly not toddlers, but this is an unexpected. Well, I'm going to need a minute to process second and third. debates that parents are going to have when they see this transcript

and it's like, Daddy told me I could eat cookies all night. Oh, man. It's going to be a complete dark. So we do see a lot of stories where parents are worried about what happened at school. Kids don't like sharing what happened in school.

So one of the first times when we actually tried the toy, the kid was very shy. And after the shyness broke, the kids started talking about what happened at school and start showing everything around to the dino. And the parents were so, happy actually to see the kids tried opening up about what happened at

school because that's a big fear parents have. So we do believe it helps parents and kids get closer together and bridge that communication gap during the early days. So a lot of the video that we saw was about, you know, ramping this up so quickly, getting in

these orders and then having to sort of build a team and figure it out, build a factory. How many dinos have you actually sold so far and like give us the roadmap here? Like, by this Christmas, how many dynos are you expecting to move? Yeah, and are you building them in China and your startup is going to like seize up if this doesn't get cleared up with China and tariffs? I hate to make everything about politics, but where are you making them?

Yeah, it's actually a crazy story. So last year, we were gearing up for Christmas and the sales blew up a few weeks before Christmas. So we were like, okay, we have to deliver all these dynos. There were some great stories, like some grandparents bought these dynos for all their grandkids. There was one grandma who bought this as the only prison for her grandson, who she was seeing after months. So we were like, okay, we have to deliver. And we started handmaking all these dynos in SF, in San Francisco. So all these dynos that we delivered so far, 1,200 of these, they were all hand assembled, finally, in SF.

But we are definitely moving to China. The good thing for us is we are not tied to China. We know we want to do it somewhere in Asia or outside of. of US to bring down the cost of manufacturing. The big companies, big companies, they're freaking out because their whole supply chain is messed up now. For us, we do have the flexibility

to produce some part of it in the US, some part outside of US. And at the end of the day, the hardware itself is actually not that expensive. It's microcontroller speaker microphone. I want to see Dino in action. Yes. Let me wake up Dino first.

All right. Dino, wake up. All right. I am waking. up. Just pre-in to be, Jason. A few seconds for me to get ready. Hey, Jason. Monday mornings are perfect for podcast fun. Ready to make startup magic?

You know it, Dinah. How are you doing today? Are you happy to be on our show? Oh, yeah, Jason. I'm super happy. This is the coolest show in the universe. We think so. Hey, Dinah, what do you think about current changes in the stock market?

Whoa, Jason. The stock market's bouncing like a kangaroo on a trampoline. I'm sorry. All right. Dino, can you do the alphabet for me backwards? Z-T-X-O-V-U-T-S-R-Q-P-O-N.

Well, congrats. You passed your roadside sobriety test. Yeah, that was half of it. That's better than my kids can do. Yeah.

All right, so, Fatin, like, what's the average query? Lon and I, clearly are not the target market here, but do kids ask,

like, go to sleep? How are you doing? Did they wait for the down to ask them questions? I'm curious, like, how these conversations

are bubbling up in the field. Yeah, so we do see storytelling as a big feature that kids always gravitated towards. Like, kids love stories. Kid will ask parents to make up stories about XYZ and parents also get tired at some point, right? Like, one of the first stories, a story request I saw was the kid was saying, Dad, can you tell me a story about a volcano, a pizza, and a squirrel? That was like, there's no way I can make a story about that, right? But she ended up asking Dinah about a story,

and Diana just somehow made a crazy story about a squirrel using the pizza as a flying vehicle and went around the volcano. I was like, this is so creative. So that's how kids usually play with Dino. And we have the phone app where parents can set up different modes or focuses. Some kids gravitated more towards playing games like guessing games or 20 questions or they play trivia games, quizzes. So it depends on each kid and that's the best thing about Dino.

It's very personalized. It adapts to your needs. And it becomes like a friend who is very understanding, very considerate. It feels like the perfect. I have to let you go in a second, but I have to ask one more question before we do. So Dino costs $200 bucks today. It's connected to the internet. It's running queries. AI is not that cheap to kind of run. And so I'm really curious how you picked the price point, how much margin there is. And if my kid loved Dino and used every day for three years. How much money would you end up losing? So we do believe the cost of token is going to go down drastically and it's actually already going down. So we're not too worried about token cost. If you think about it, the conversation

tokens are much lower than, let's say, code generation where you just done paragraphs, such paragraphs. It's like a few lines at the time. So token cost is actually going to keep going down and we are going to switch to open source models in a couple of months, actually. So token cost-wise, we are not too worried. And on top of that, there will be a subscription cost to cover up more of these costs. And the subscription actually won't limit the conversational abilities. It's going to provide extra capabilities to parents through the phone app. Like being able to ask Dino anything through the phone app, like, hey, is there anything I should be worried about or what are my kid interested in these days? So,

it's actually a sidekick for the parent through the parenting app. Yeah, it's kind of like spying on your kid's imaginary friend, but you keep saying things that make me more curious. So which open source models are you looking at? And then are you going to self-host those or just rent CPU or GPU time via a regular cloud provider? So we haven't actually finalized which open source models we are going to use. There's like one new groundbreaking open source model coming up every single week, every month.

So we're in the talks right now with a couple of these. So for now we use some of the leading API providers But the good thing is we have built this platform Where we can hot swap these models really fast So no matter whoever wins the AI race We end up winning as well We are not competing with OpenAI

We are not competing with any of these big cloud providers We are a separate platform on our own All right and if you want to take a look at this It's magicaltoys.com Take a look at Dino If you have a child maybe buy them one I would but my spouse doesn't like having microphones in the house

So I'm out of the customer base, but I think this is fantastic. Patine, thank you so much. We'll have you back on. It's the new versions out. Yeah. Thank you for having me. Patinth, thanks for coming. Great to have you.

That's it for today's Twist. Thanks so much, everybody for being here for at Jason Calicatus. I'm at Lonz, Lon Harris. And I'm at Alex. You can find us on TWA startups on X. We're over on Reddit. We have social platforms everywhere. And to wrap up this episode of Twist, we're going to throw it to an interview that I did with Cloud Neuro,

a company that wants to help other companies, manage and understand their SaaS spend. We get into SaaS fatigue lawn and how to go against kind of the enterprise sales market of today. It's quite a lot of fun. We are everywhere you get your podcast. We go live three times a week, Monday, Wednesday, Friday, roughly noon Texas time, one p.m.

East Coast time. We adore you. We're back on Wednesday with Founder Friday's Final Four. Get excited to everybody. We'll see you then. Bye. Hey, everybody. Welcome back to Twist. This is Alex. And today we are talking about taking on the SaaS sprawl. Yes. Now, in the

last 10 to 15 years, software removed from the on-prem world into the cloud and became a managed service, thus bringing us software as a service. Everyone loves this, everyone uses this business model. Pretty much every startup you know is predicated on the SaaS approach to selling things. But a lot of companies bought more software than they really needed, or maybe they have seats that are not being filled. This has created a niche in the world for companies to help other companies manage their software spend. It's a big category. It really matters. And that's why I'm very very excited today. Talk to Shom Kumar, the co-founder and CEO of Cloud Neuro, a startup that is working on taking on this exact problem, and I want to know how he's working in a competitive

market and just how big SaaS fatigue really is. So please welcome to the show. It's Sean Kumar. Sean, hey. Hey, Alex, thank you so much. Thanks for having me. Oh, my pleasure, man. I always love learning about what's going on actually out there in the world. And, you know, Cloud Neuro, because it's sitting there talking to companies about how to save them money, I presume you have a pretty good finger on the pulse for the state of the enterprise buyer and just the health of the software economy in general. But before we get to all that, just tell me what Cloud Neuro does to make sure that I'm not talking out of the side of my mouth here. No, thank you. That's a great question. So on an average, enterprise companies are using over 600 plus SaaS applications.

Uses are growing day by day, you know, in some cases, 10%, but in some cases, they're 27%. So there is more and more SaaS or a mix of SaaS and platform. And now with the AI, there is more, more SaaS is coming. And this growth is creating a lot of overhead in terms of a lot of misuse, accountability gap, and so forth. So what Cloud Nuro has done, we have built a set of tools, basically we sits on top of these platforms or SaaS and provide a complete accountability across all these.

What it means, it provides a single pane of glass view of the entire CloudSAS landscape. It provides where the money is being spent, where who is being, why they are being spent, so that every dollar that is spent on this Cloud and SaaS are being tracked through CloudNover. So this is precisely what CloudNus is doing. Okay, so to make sure I'm fully tracking this, essentially, CloudDurl will come in, look at a company's devices, see what software they're using, figure out what they're paying for. Do you guys track, it's kind of like gray market SaaS?

Like if I'm an employee and I buy something and put it on my computer, but it's not centrally sourced. Can you guys also find that when you're kind of scanning through a corporation's software? Exactly, Alex. So that's where it begins. It begins with something called auto discovery. So we continuously discover different cloud and SaaS that is being used.

We have different discovery channels through that. we keep looking at what are being used and we start flagging back to the users, the various anomalies of SaaS, those are not supposed to be used versus being used, and if it is being used, then how it is being used and all those. So yeah, that's the core functions of this product. Thinking about controlling SaaS inside of a company, one thing I've heard from a lot of startup founders is that it's a little bit harder to sell these days because people are being a little bit more tough on price or not buying as many seats. forth. But if you already have a software product, you might end up with kind of more licenses

than you're really using. So as part of the auto discovery feature of Cloud Neuro, can you also find out like, okay, we're paying for 20 Slack seats, we're using 8? Can it actually tell where you're probably overspending compared to usage? Yes, absolutely. In fact, we help companies. So many times, it is taken in a way that we are minimizing or cutting the cost. In fact, we are helping to sell more, if you think that way. For a customer, take an example of Salesforce. To one customer, Salesforce may have sold X number of seats of subscription for sales, cloud, service cloud, a marketing cloud.

Many times, those are not used the way it was sold. But now, a sales agent wants to sell Salesforce agents and customer don't have budget. So what we as a cloud noodle, we do, we keep tab on those and we tell. them, all right, you cut these because you're not using. Instead of you retrofit these dollars to use your buy some other agents. So every dollar that is being spent on these SaaS, we are ensuring that those are rightly used, effectively use, and both customer and the product vendor, they all win through that. In a way, what Cloud Neuro does is provide a lot of visibility to company owners to ensure they're not wasting their money. What it does to companies that sell

software is uncover revenue that they might have to lose. And so I kind of wondered, do people ever call you up and say, like, please stop because you just cost me, you know, X number of seats of my software over, you know, Y number of months? You're kind of like the police, if you will, of SaaS spent. A rather good police, I would say. Our job is to help these companies where they really have a purpose for every dollar being spent. In addition to that, many time in most of these cases of SaaS deployment,

we always find that there is an anomaly of users. The users who are sitting there, they are not supposed to be sitting there. The users, their entitlements are at a different scale, what it should be there. So what happens that in one case, we found a company which has 100 SaaS users and 18 of them were provision as an

administrator right so so these are the huge SOD compliance issue

so and when you have hundreds of SaaS there is no easy way

that you can go and look at each one of those how their entitlements

are and this is where how we help and ultimately the cost

optimization is a win for everyone it's not the cost reduction versus

you you're giving an opportunity for each of the vendor to bring the best product that customer want to use versus what you sold and customers not using that anywhere they will find someday and will stop using. Yes, and my guess here is that if you can find someone perhaps overspending on a certain product, save them that money, and make that relationship with that vendor healthier, you probably reduce churn in the long run because the actual product in question now costs less than it would

have, if you hadn't uncovered, the overspend or the miss spend, you might say. Exactly. In fact, just this morning, you will be surprised. We were speaking with a customer and one of the top university in Chicago and their worries

about the agent team. Their spends are huge. People are using everybody is using this Gen AI and they are just doing query and ultimately there is the cost, you know, there's is a flag that's going every time.

So for them to know who is using, what they are using, ultimately, even each AI application is a SaaS, is a subscription. So that needs to be managed, right? Yes. Well, I want to get to AI in a second, but to me, you sit as a company in between two competing forces. On one hand, everyone wants to save money, which to me would imply that demand for what Cloud Neuro does very high.

On the other hand, people do have, and I've heard this from lots of folks over the couple of years, a little bit of SaaS fatigue. They don't want to buy yet another thing. And so it seems like you have kind of a headwind and a tailwind, Sean. How do those balance out when it comes to how the company grows and approaches the market? We go to the customer to build them the real foundation for managing cloud and SaaS. There is already a budget set aside for certain SaaS spend. And we are a real foundation. our budget do come from optimizing those sets. So we are not creating a new budget.

What we are offering customers more than what they already have and just piggyback on those and helping them optimize. So that's the real sense. The fatigue is in terms of customers do have, many customers, has many, many competitive SaaS, many, you know, same product. doing multiple things and multiple products doing the same thing. So as long as we help them bring

that visibility and control in their landscape, I think we have enormous business. And by the way, just so you know, only 10% of this entire SaaS is right now being managed through a centralized

this kind of platform. And we're talking about $300 billion total SaaS has spent this year alone. So they're a massive market. Yes, and to be clear, you guys do more than just cost control.

that's just the part that stood out to me the most. But also, I think you guys have a portal that you offered to let employees find approved, corporate-approved SaaS applications as well. So it's also like a way to get people to write things they should be using versus stuff that they shouldn't. Yes, precisely. So that's our, you know, the workflow. So the employee comes to our website or rather Cloud Noodle. They can, with catalog, they can select whatever SaaS they want to use or,

or recommend to buy. And through that, then it goes as a part of workflow and a manager approves, and ultimately, you know, they have everything what they need through a proper channel. Okay. I want to go to AI now because we can't in 2025 talk about anything without spending some part of it on AI. Now, in this case, I was going through the Cloud Neurosite, and you guys have a product called AI Custodian, which you says optimizes AI spend across hypers and AI applications.

So when I see a company add an AI-focused feature lately, I'm always very curious. Is this something that you're adding to the platform because you expect in time that your customers are going to want this? Or did the market tell you, hey, John, we need this. Can you build this for us and add it into what Cloud Neuro currently offers? I'm curious. You know, before this JNAI came, we founded this company. And we, for the foundation was the long term, we were.

want to be, we are a data company. Primarily, we, we subscribes the data usage. And we always want it to be at a stage where we tell companies what exactly the software, how much this would be spending, how much this would be spending in different categories of software over a period of time. This is our visit. So company don't have to guess and figure it out what their spending should be because every other spending we know and this spending is to.

totally to be guessed by some IT operator. So that's the bottom one. So AI is in our foundation everywhere. As we design our system, right at every layers, we have good of AI, whether it is orchestration, whether it is data integration, it's a mix of AI everywhere. But as more Gen AI did come, we started using Gen AI to primarily get more user, better user interaction and providing the usage predictions.

Like in our platform, we also, besides SaaS, we also do cloud optimization. So all the cloud usage and how the cloud is being provisioned with a user of AI, VES, startup. We also have created our own prompt, or rather optimize our prompt, to identify the non-human identity because identity is extremely. important because it's a user, non-user, how identity is being used across the system. That very much defines how security you are, how cost-effective you are. So for us, using those Gen AI, bringing those identities, visibility is and we use.

But customers, did they come to you and say, hey, listen, we really need better visibility into our own AI spend? Can you ensure that the cloud neural platform handles that as well? Or did you guys just kind of see AI rising? No, so much. So now the recent task is all about agent governance. Like we have a SaaS sprawl. Now the agent sprawls are happening.

Like every other SaaS providers are throwing one agent to the customer. So customer now really want to know where is my agents, where are the agents, how much I am spending. So this is the new. And we are so excited because our platform is ready. It's just a matter of hooking those up and ensuring that the isn't. and governance. One thing that I would love to know from you guys, if you ever built a page for this on your website, is like, your average new customer you get, let's say they spend X on SaaS, X on Cloud.

I'm curious what the comparative spend is on Agentic AI or maybe just Gen AI in general, because I presume if you look at that chart, if we did it over time, it would be like, you know, SaaS, cloud, and then recently be like AI. But I'm curious how they compare. And if SaaS spend is near to the average cloud bill or cloud bills near to the average Gen AI bill. Because for me, for where I sit, Shamm, I've no idea. It'll be fun to see those ratios change as the industry adopts more AI. Absolutely. And yeah, it also depends upon industry to industry.

Yes. Obviously, right now, the cloud spend is the number one. And then comes the top SaaS. This is where the custodian term, what we used, what we found in our research. And by the way, we co-founders, we have spent decades on just SaaS. Like, this is the one thing and only thing we know the best, if I can tell you. So, what we have found in our pattern that 80% of the spend of the total SaaS comes from top 20% of the providers. And these are SS4, Service Now, Office 365, and like like, right,

a workday and others. And so the pain is also. coming from these top SaaS providers. So first, the biggest spend comes from mostly infrastructure, like public cloud, and then the rest is coming from these top SaaS providers. And yeah, AI is catching up, right? AI spending is growing, and very soon, there will be only maybe AI spend left.

I'm curious about how the company is growing. I know you guys are pretty early stage still, but talk to me about 2024 versus 2025. What are the goals for this year? how much are you going to grow? Yeah, so 2024, we have grown 400%.

Yeah, 4X. And this year, also 2025, we are going to go at minimum 3x, 2 to 3x is our growth. Compared to Zylo, again, Zylo and other companies, they are

doing great SaaS management and kudos to them what they're doing. But we are creating a new space in that, where our platform, is capable of managing public cloud, SaaS, and AI, and together, and bring one accountability. You don't want to buy for each one of them one product. Because then you have more SaaS sprawl,

which is what we're trying to fix. Exactly. Exactly. So that's the bottom of when we build this product, the root, because we understood this problem better than anyone. We knew that this problem will be there. And then on the top that you build this product, you build that accountability layer so that everyone, a CFO knows what where the spending

you. CIO knows why this spending is. Likewise, there is one very unique thing that we are building, and I want to announce today, I was holding it. For public cloud, there was something called chargeback and cost allocation. So in a bigger organization, the central organization do purchase and they distribute and then collect the cost and payback. But this was never thought for the SaaS. But big SaaS like Salesforce or set up security product or service now, larger organizations, we have a larger profit center, they all deal with this problem. So we, Cloud Nuero, we are building the cost allocation and charge back for the SaaS applications. So that every dollar bought, used to reconcile is all coming from one platform and everybody

knows of this. Okay. So as not a CFO or a CIO, I'm going to repeat that back to you to make sure that I'm fully tracking. Public cloud infraspend is so large that it internally gets allocated to the right cost center at a company that uses Azure or AWS, whatever. You guys want to apply the same idea to SaaS spend so that way you can more directly apportion cost internally to the right cost center and not have it be just here's our Salesforce spend. It's one bucket, even though everyone kind of pulls from it. Yes, exactly. Everyone pushed from it, push differently.

Marketing has a different need for sales force versus sales versus internal app. But the cost should be different because each one has their own profit centers. So not knowing exactly what they are spending and they all have to go to a spread seat and do back and forth, spend hours and hours to figure it out. Now they will do in a real time to Cloud Nooro where their budget is in a fingertips. I'm very glad to hear that. And I have to let you go in a second, but I want to make sure we talk about my favorite topic with Midwest founders, which is you are building in Chicago, the Chicago area, I should say.

And we were talking before the show about, you know, where you've lived around the world and you've ended up in one of my favorite places in the entire, entire earth, frankly. So just before we go, what's it like building in the Chicago area today? How is the talent pool? I don't know, just what's it like? I mean, you're not in San Francisco. Thank God. Someone finally.

Yeah, no, thank you. Alex, Chicago is one of the most beautiful city, and I invite everyone to come this spring and summer and fall. You will not find a better city than Chicago. Just don't go in the winter. That's all we're saying. Don't go. Yeah, if you can, want to ski, please do come in the winter too. And then enormous amount of talent pool, the beautiful places, they're enormous universities here. So that's definitely not a problem. And for us, for me, I'm privileged to be living in suburb. called Naperville, and Naperville, one of the top 10 city to live in the country. So, always they are in the top 10, and obviously one of the best city to live in. It's an entrepreneur city. There's a lot of companies, a lot of startup that comes from this small town in Naperville, which is a western suburb.

So, yeah, this is Chicago is the best place to start business. Right now, Chicago is too much hype about the quantum. So there's a lot of any city of the city of Chicago. That's so, that's interesting because people often forget that it is the Seattle area that is the real hub of the global cloud industry. It's not the Bay Area. It would be great. And just like biotech is, you know, kind of a Boston thing. It'd be fun if quantum became like the Chicago marquee.

So that way we would have another city with a specific focus to have its own, I don't know, flavor or take on the world of startups. Anyways, Sean, two things. One, what is the URL for the company? And two, what is a role that you are having a hard time hiring for? I want to make sure everyone knows what talent you need. Yeah, thank you. So the company URL is cloudneuro.a. Very simple, cloud, nuro.org, cloud nero.

And we are really looking to hire one sales rep. So we have huge pipe generated. We need somebody to help us close. So preferably in either Chicago and Midwest or in the West Coast. If we have a sales grab, that will be amazing. And best way to get a hold of you, email, Twitter, what's your favorite? Yeah, email sham.comar at cloudno-do.com. Simple enough.

All right. Well, Sean, thank you so much. And I want to talk to you in like six to nine months to see how this year's growth is coming out because I think as everyone looks to keep a tabs on their SaaS bend, you're going to have a really good year. Thanks, man. I appreciate it. Thank you, sir. Have a good year.

Thanks, Alex.