This Week in Startups - AI demos + Jason invests $25K live! | E1789

Episode Date: August 8, 2023

This Week in Startups is brought to you by… Lemon.io - Hire pre-vetted remote developers, get 15% off your first 4 weeks of developer time at https://Lemon.io/twist Embroker. The Embroker Startup Insurance Program helps startups secure the most important types of insurance at a lower cost and with less hassle. Save up to 20% off of traditional insurance today at Embroker.com/twist. While you’re there, get an extra 10% off using offer code TWIST. House of Macadamias is the next big health trend! Get 20% off your first purchase and a free box of Namibian Sea Salted Macadamia Nuts at houseofmacadamias.com/twist by using code twist20! * Today’s show: First up, Sunny Madra joins Jason to demo some new AI tools, including auto-generated articles, social clips, and ChatGPT’s newest feature. (3:50) Then, Jason sees pitches from three recent Founder University graduates and invests $25K in one of their startups! (40:33) Time stamps: (00:00) Sunny joins Jason (2:20) The pace of progress in the industry and influx of AI startups (3:50) Sunny demos byword.ai (9:58) Lemon.io - Get 15% off your first 4 weeks of developer time at https://Lemon.io/twist (11:19) Building a company today and improving LLMs (17:21) Sunny demos Submagic (23:53) Embroker - Use code TWIST to get an extra 10% off insurance at https://Embroker.com/twist (25:09) Sunny demos ChatGPT’s new Custom Instructions feature (29:40) TWiST's "all-star" summer and brief TWiST history (39:13) House of Macadamias - Get 20% off and a free box of Namibian Sea Salted Macadamia Nuts at https://houseofmacadamias.com/twist by using code TWIST20 (40:33) Jason and Kelly kick off another $25K startup pitch competition! (44:13) Rami pitches Querio.ai (49:18) Batuhan pitches BeforeSunset.ai (59:54) Allen Pitches Semanttic (1:07:03) Kelly and Jason share thoughts, then Jason picks a startup! (1:11:34) Q&A with Jason * Check out byword_: https://byword.ai/ Check out Submagic: https://www.submagic.co Check out Querio.AI: https://www.querio.ai Check out BeforeSunset.AI: https://www.beforesunset.ai Check out Semanttic: https://www.semanttic.com/ Follow Sunny: https://twitter.com/sundeep * Read LAUNCH Fund 4 Deal Memo: https://www.launch.co/four Apply for Funding: https://www.launch.co/apply Buy ANGEL: https://www.angelthebook.com Great recent interviews: Steve Huffman, Brian Chesky, Aaron Levie, Sophia Amoruso, Reid Hoffman, Frank Slootman, Billy McFarland, PrayingForExits, Jenny Lefcourt Check out Jason’s suite of newsletters: https://substack.com/@calacanis * Follow Jason: Twitter: https://twitter.com/jason Instagram: https://www.instagram.com/jason LinkedIn: https://www.linkedin.com/in/jasoncalacanis * Follow TWiST: Substack: https://twistartups.substack.com Twitter: https://twitter.com/TWiStartups YouTube: https://www.youtube.com/thisweekin * Subscribe to the Founder University Podcast: https://www.founder.university/podcast

you know, like Reddit or this or Twitter. When you can have anonymity in the system, you know, all kinds of Fugasey stuff can come along. You can pay people to get your, you know, edits through, and then other folks are incredibly principled, and they will lock pages, and they will fight to the last dying death to revert stuff, right? And anytime you try to appeal or do something,

like, they just kind of admonish you for how dare you try to correct an error on the page of your bottom. There's no resolution. And that's why people who, you know, anybody who's notable looks at Wikipedia and they're like, yeah. They trust it in a different way, right? Yeah. Yeah. Then people who don't have a, I wish my page got deleted.

I tried to have it deleted a couple of times. It's not notable. And that didn't work. This weekend startups is brought to you by lemon.io. Need to speed up your product development without draining your budget. Hire vetted engineers from Europe at lemon.io. Go to lemon.io slash twist to get 15% off for the first four weeks. Embroker's startup insurance program helps startup secure the most important types of insurance

at a lower cost and with less hassle. Save up the 20% off of traditional insurance today at imbroker.com slash twist. While you're there, get an extra 10% off using offer code twist. And House of Macadamia's is the next big health trend. get a free box of Namibian sea-salted macadamia nuts at House ofmacadamias.com slash twist. Plus, get an extra 20% off your order by using code Twist 20. All right, everybody, welcome back to this weekend startups. It's Monday.

Sunny Madra is back. Do some AI demos and talk about the state of AI. We'd like to kick the week off with that, as you all know. And then I got three startups from our founding university program. we're going to pitch me live on air and I'm going to invest $25,000 in one of them. And all three are AI-based products. So you can get you see some great AI demos. Sunny, welcome back to the show.

Good to be back. Yeah, I'm excited. All right. It's August and the industry is not slowing down. It feels like things are accelerating. I can't imagine what the fall is going to be like. I mean, the amount of effort people are putting into iterating on these AI models and building new experiences for consumers is unlike anything I've ever seen.

Yeah, I saw a tweet the other day that there's 6,500 companies in AI right now. That feels directionally correct, and I have to say, one of the things that I'm super amazed at is how much product people are building given the resources they have. In other words, two or three person teams are building stuff that is. wicked. And so reminds me just like cloud computing, used to be like, hey, when can you get this up and running?

And then all of a sudden they'd have an app up and running and a site up and running within 30 to 60 days. And you thought, hey, it wasn't it take three months to set up your co-location facility and order to the servers and do all that. Now you're going to be up and running in days to weeks, as opposed to months or quarters. It's really, really fast. But I think it's awesome.

The pace is good. And I think that's a good intro into our demos. Yeah, let's get started on the demos. And I showed you some of the stuff I'm working at inside on without talking about it. It feels like I got something? Do I got something? Yeah, you've built something that, you know, is representative of two or three companies today, but then haven't put that together. And I think it's like the next iteration of it. So I really, really like it. I'm very excited for it. All right. So let's do our first demo here. Let's do our first demo. All right.

So this one, J-Cal, I think you'll have a lot to say about it because it sort of reminds me of a lot of what you were doing a little bit with Mahalo. Human power search. Yeah. But, you know, kind of building the, you were building the pages. And so in this case, I asked it to build a, like an SEO page for the All In Summit. And I gave it some keywords, which, you know, I can pull up in a second. and you can see here it came up with a little graphic and an overview. So I just said, All In Summit, Royce Hall, give it, you know, all the host names, and it, you know, did all this on its own.

Amazing. I thought that was really good. And this is, this particular one was designed to be keyword optimized for SEO. And so I thought that was, you know, really exciting. And it took me less than five minutes to do this. And you can kind of see the settings here. you can add custom prompts to it, word count, tone of voice,

all the stuff that is becoming obvious now, or, you know, table stakes. Exactly. They give you also, and I'm just clicking for those folks listening, I'm just clicking on like the socials button. And so they also give you a little previews that you can dump into Instagram,

LinkedIn, different tweets that you can do. Super powerful. And so they have everything from like single article generators, you know, which I'm clicking on here, which is what I used, you can also do batch article generation. Oh, no.

I mean, the pollution of the web now is going to be insane. I can see this already because I'm starting to get AI reply guys on my Twitter. So people reply with something that makes a little bit of sense

and then it feels like, okay, either a person who barely speaks English or an AI bot is replying to this. You know, like when you would have your Facebook groups get infiltrated by,

you know, Manila spammers, right? And they were being told to put links and just, you know, reply to anything just to get clicks. It reminds me of that where people who had English as a second language and were trying to make some contextual reply. And that's, I think, what this is going to do. This is going to pollute the heck out of the web. I hate this kind of automated batch generation.

I do like this as a starting point for a website builder to kind of give them ideas, right? And then take it from there. Yeah. Or like, you know, I'm considering using it for definitive. Like, you know, we have a bunch of articles that we want written on certain topics, and it kind of gives us the framework for those things. And, you know, in these custom formats, you can have it be programmatic or pull things from URLs and things like that as well.

So, I mean. So, yeah, you could say, write a story. You could point to, let's say, I don't know, the New York Times homepage. And you could say, go to the New York Times. com, pick a story, and rewrite it essentially, and, or write a summary of, I don't know, I'm trying to think of something that gets public. Oh, a Mr. Beast video. Here's the Mr. Beast video of the week, write up an article about it. So every time he posts on Saturdays, the latest video, write an article about it and post it to my Mr. Beast blog. And it's just kind of feels like

automated spam, but I guess if done right, it could be high quality content to navigate the Mr. Beast video. So we'll see. It's a whole new world here. I mean, I think what this will force is the search engines to get a lot better. And, you know, I think this is good for your website, but I think if you're using it for kind of spam purposes, I think the, you know, we're going to see technology that can start to detect that these things were written by AIs.

But I, you know, you've been doing this a lot longer. So I am really, because your reaction is pretty, you know, lukewarm to negative on these. Yeah, so what I think is because the quality is like generic, I think it's enough to confuse the algorithm into thinking it's a human, but not good enough to provide value for humans. So it's kind of like falling between, I think it's going to beat Google and Bing and they're going to have a hard time dealing with this and figuring out if it's computer generated or by a human. And then I think conversely, I think humans are going to get to this and feel like, one of my, is it. Is this just filler calories? So it reminds me of like, you know, you eat some organic strawberries, and they're small and they're delicious and they're dense with flavor,

but they're imperfect, you know, kind of human-ish. And then they make these factory ones, you know, that are giant strawberries that are like the size of like a small apple or a lemon. And you're like, that's too big for a strawberry. It has no flavor. I think that's kind of this artificial, you know, web of, content that we're about to create that's going to destroy the open web. I think this is going to flood and destroy the web if it's not contained.

Now, if it got good enough that it was providing actual value, great. But when I look at that, you know, All In Summit review or any of these, it feels like filler content. So I get it for brainstorming. And it certainly will get you to like second base. It gets you 60% of the way there. But man, you really do need to have. have a human polish this,

which I think means the great writers will become even more valuable in this world. Because average to good content is going to be everywhere. It's going to flood everything. And the automated nature of this is also troubling. That to me is the,

and the batch stuff. So you could say, rewrite the entire Wikipedia. I could rewrite the entire Wikipedia right now and say take every section of the Wikipedia, rewrite every section, and do it in this tone, a funny tone, a clever tone,

a more professional, a less professional, you know, whatever it is. And then how does the internet deal with that, right? And supposedly it's happening on Reddit too, that there are Reddit bots that are now posting comments. Okay, listen, you got an idea for a tech startup. Great. You think you want to change the world.

You think you got this. This is the one. Well, you've got that same problem that we all do. You don't have an engineer or you don't have enough engineers to make this happen. And you need product velocity. You need to go fast. And how are you going to go fast?

And how are you going to control your burn rate? if you got no engineers, well, what if you had a partner who could provide you with more than a thousand on-demand developers? And those developers were all vetted, experience, result-oriented, and passionate about startups and building great products. Well, what if they also charge competitive rates? Does this all sound too good to be true? I know it does. Well, then you need to head to lemon. dot I.O because your dreams have come true right now, startups, choose lemon.io, because they only offer handpicked developers with three or more years of experience and ones that have strong portfolios. Only 1% of candidates who apply to work at lemon.io actually get accepted.

And if anything goes wrong, Lemon will get you a replacement ASAP. A couple of great launch founders have worked with Lemon.io, people in our portfolio, and they have had great experiences. So I want you to learn more at lemon.io slash twist. And when you go there, you're going to find your perfect developer or an entire tech team. And you're going to do that in 48 hours or less. And Twist listeners get 15% off the first for a week. I want you to stop burning money.

I want you to hire developers smarter and faster. Visit lemon.io slash twist and give it a shot. If you're starting a company, J.Cal, right? And you're obviously supporting this all the time. This used to take going out and getting some combination of marketing and PR firms and all that. And if you're a two-person startup and writing is not your,

you know, forte, but your great engineers. Like I think back to when I was starting my first company, it's like I'd love to have something like this, just to give me the basic framework. You know, we live in a weird society where we have to take, you know, concepts that are filtered down and expand them to make us look, you know, more professional. So that's, and that's why we have to do it. Like, if I could write three bullets and put it on my website and say, come use, you know,

definitive or whatever, but like, you know, a big. I think as a, yeah, so I agree with you as a brainstorming tool starting you off on second base. Yeah. Yeah, I consider it like the templates at Squarespace. Yeah. What was really amazing about the templates that Squarespace created in the early days was they were better than what you could pay a designer to me. Now, I don't mean not to diminish any designers, but they figured out a way to hire the best designers, build this collection of templates. And so I would look at a Squarespace website and they still do today.

And then I'd have a web developer build something. I would like, Squarespace template's better. I literally just went through this because somebody on my team was like, I'll just make this and, you know, Webflow, whatever. I'll, you know, handcrafted, I'll do a no-code solution. I was like, oh, that's all great. And then I looked at the design and I was like, can you just use a Squarespace template and make something gorgeous? And so, and I love all those no-code platforms.

I think they're great for startups, etc. But that's what this is like, doesn't have yet, where I don't think the output is there yet, you know. And so somebody's got to make the output really crisp. But sure, as a starting point, it's like an okay template to start with. like a, yes, a five or six as a template as opposed to a nine or a ten. I think it'll get to a nine or a ten, six months. We'll come back to this one.

I mean, I think that'll be, when you can show me an article written by this and then a professional writer writes it and then I can't tell the difference, which is what I've been trying to figure out in my projects is, you know, at what point can't I tell the work being done by humans versus the work being done by machines? Yeah. And so six months, okay. Early 2024, early 2024, you won't be able to tell. You won't be able to tell.

So you think this is going to go from a five or six out of ten to an eight, nine or ten out of ten in terms of content. Between eight and nine. Yeah. Wow. That's a big jump. You know, these, and I think these last couple of jumps are the hardest. So I'm very interesting. What do you think is going to cause that?

Because people have been saying that maybe chat chip EPD4 is deprecating and it's giving worse results as it ingests more. Maybe it's getting more garbage in there or something. What do you think is going to drive this, you know, the next 20%. So there's two ways that LLMs are getting better right now. One is more training data, which I think is tougher because of, you know, everyone's much more aware. We've talked about it a few times on the pod. The second is the expansion of the parameter space.

And what that particularly means is when you give an LLM more parameters, it's like giving a brain more neurons. That's the best analogy. And so as the parameter spaces increase, And we already see in the open source, we have seven billion parameter open source models you can run on your Mac, right? We're going to start to see the open source expansion of models with more parameters, which means they're going to get smarter and smarter, not having to pull in more data. Yeah. So we're going to see that.

Yeah. Yeah. I mean, the hallucinating is a thing. I saw this week, like, people are getting, somebody was suing because an AI was hallucinating. on a real person's bio. So you ask, and this happens with mine. You ask it, who's Jason Galacennyson? You know, get 60% correct, 40% is going to be a complete hallucination about what somebody

in my field might have. Like, it might think I went to Stanford because people in my circle went to Stanford kind of thing. But you know what? I've already got a dry run for this, which is the Wikipedia. If you're, the biggest problem on Wikipedia is bios of living persons, BLPs. If you're a living person, you know your bio is complete utter rubbish and it's driven. by all your haters or people with an axe to grind,

anybody you fired or you crushed in business, they basically spend all day long reverting changes and with your, sorry with first, messing with your, uh, bio on Wikipedia. And like my bio on Wikipedia has nothing to do with all the great things I've done

in my career. It's like little weird fights I got into or obscure, you know, uh, projects that I did. Yeah. None of the like major stuff is on there. And I gave up even trying to change it.

I think this is where the, LOMs are going to have challenges. Isn't the core other fact that there's only like really like a hundred or 200 people that ultimately all that bubbles up into? Like anybody can go and edit, but the real kind of masters of the universe are just a few hundred people or less. There are secret cobbles of editors who you can either pay or are just very principled. So there's a, you know, like any other anonymous social network, you know, like Reddit or this or Twitter. when you can have anonymity in the system

you know all kinds of Fogacy stuff can come along you can pay people to get your you know edits through and then other folks are incredibly principled and they will lock pages and they will fight to the last dying death to revert stuff right and anytime you try to appeal or do something like they just kind of admonish you for how dare you try to correct an error on the page of your bottom they just there's no resolution

And that's why people who, you know, anybody who's notable looks at Wikipedia and they're like, yeah. They trust it in a different way, right? Yeah, yeah. Then people who don't have a, I wish my page got deleted. I tried to have it deleted a couple of times.

It's not notable. And that didn't work. So what else you got for us? Okay, next one. This one is, you know, kind of inspired by, so I'm making sure of my windows right here, inspired by, you know,

for me, there's the guys that take the all in pod and then turn it into like these TikToks. And so, you know, earlier today on our chat thread, I asked Nick for a little video. And all you do is you upload the video. It'll transcribe it. You can make some design choices around fonts and colors. And then it'll even give you a description for posting. So this is that little short little short video.

So I'm going to play it. It's only like 15 seconds long, all right? Yep. And the company we're reviewing here is submagic.com. The thing I love most about this is like, we have this Oppenheimer film comes out, all of this energy at the same time of people who are just really stoked, do material science, to do basic science to figure out. But, you know, what we're seeing here is a typical TikTok stories type. TikTok or, yeah, what do they call them on YouTube? Shorts. And so it's like a shorts or a TikTok video, but it puts the subs on it magically.

and you can pick the colors, etc. So, you know, that would be something that would normally take, you know, an editor, 20 minutes to do properly, maybe even 30 minutes. So the fact that you can do this quickly is pretty magical. And the ability, I get pitched every day by people who want to for $100 a clip, $50 a clip, $20 an hour, $30 bucks an hour, $50 an hour,

want to make clips of stuff. and I think the level of clips they're making, which is to say they don't understand context and put a ton of time into them and kind of are of minimal impact. The ones that make great impact is where they understand the context and they put supplemental video in

and they edit it and they really are thoughtful about constructing a message. So again, I think this is like going to create a bunch of clutter on the internet to start and because you can just point any clip at it. And then if you could auto post it, I could just literally make a Joe Rogan clips

or a Lex Friedman clips or an all-in clips or twist clips channel and just have to be automated and just throw garbage out. Just take every section of the show. Who cares if you flood the zone with a thousand or ten thousand clips? If one of them clicks, they click. So what's the

end game then, J-Cal? Because I think we all want to consume these. Like, I actually like the all-end talk. I think I follow them. Yeah, they spend two hours and they're like college educated really gifted guys and they do a great job with Allentalk where they pick three or four great moments and they really edit it down. So, you know, they might take a 10 minute discussion and turn it into a 90 second discussion with the best moments and they really are thinking it through. So that's what this isn't doing.

So again, here we go. The brute force of AI just, you know, giving you a clip, putting subtitles on it. maybe even thinking this is an interesting discussion for some reason. You know, give it a prompt, act like a producer, find me the three most interesting moments in here. I don't know how we do that. Versus a human saying, hey, I listen to the pod, this argument over Trump indictment, this argument about this discussion of LK99 and this discussion of, you know, venture capital seed funding are the three really interesting ones. I'm going to frame them. I'm going to start with Chamoth and then I'm going to go to Jake Al. I'm going to start with Sacks.

and then cut to a, you know, whatever, a clip of Trump actually, you know, in the real world. Between those two is a lot of work. I would say it's like a three out of ten versus a ten out of ten. And it's really hard to make viral clips on TikTok and figure into, you know, that's basically like Mr. B. skill. So this one, I say there's a big gap, a big, big gap of understanding what the most important parts are. Maybe I'm wrong. Maybe.

Well, I mean, I think what you just said is what Nick and I did. So Nick gave us a little compelling clip. You're talking about Oppenheimer and you're talking about how people should be really inspired right now. And then it's a really nice clip to add, you know, into the repertoire of things that, you know, Twist puts out there, right? In terms of all your socials and everything else. We use a protocol VED that does auto captions. And it takes us an hour or two depending on the video to do really good clips that are thoughtful. But VED does save us a lot of time.

And so I think BAI tools. the piece it really needs to figure out is what is the most interesting part of the podcast. And I don't know how we could do that. Now, I know how YouTube could do it. YouTube knows when somebody rewinds a video and scrubs. So YouTube could have a big advantage here. If YouTube said, hey, these are the ones where the most interesting videos are. Or you can put comments onto, if you put a timestamp in the YouTube comments, it will create a link.

right. So if you put 505 colon 10 and we'll go to five minutes and 10 seconds, that could be very interesting because if people are calling it out and enough people are calling out a moment, that could be used as signal, right? So that's what Veed should do in these other things, is look at the comments and see if they correlate in some way with the moments in it. I think Twitter would be able to do or X will be able to do that too, right? Because, you know, you guys are posting natively there. And, well, I mean, on this week's one, even Elon chimed in on the, you know, how important is LK99, right? So you can imagine.

He did. He did reply to me. Yeah, yeah, yeah. Yeah. So that was, yeah. So, I mean, that's the kind of signal that you're looking for. So then you have two pieces, you know, to train the a ion. You have the core piece of data.

And then you have the world's reaction to that data. Yeah. Implicit or explicit. Explicit being a comment or people rewinding. Implicit is sort of the, the views dropping off, stuff like that. Dropping off, yeah, like that kind of behavior. So interesting.

I think this is again where, you know, augmenting human performance, these gains are going to be 30% this year, 30% next year, and that's double. So once again, if you are running a company right now, you need to get everybody in a room, everybody has to learn these new tools, everybody has to learn AI, or else your company is going to fall behind dramatically. Listen, I work with super early stage companies at launch, like literally year zero. They haven't even incorporated yet. And then we hit the Series A. People have thousands of dollars in MRR.

And maybe they've only raised a couple of hundred thousand before that Series A. And they don't have their insurance set up. And in fact, we recently had a great startup that didn't have DNO. And we had to really stop everything because they were having board meetings. They were making massive decisions. There were legal issues. And they didn't have the basic D&O insurance. that protects directors and officers.

So we send them right to Embroker. Embroker is business insurance built specifically for startups. A single application will help your startup get four quotes for four lines of coverage in 15 minutes. Think about that, four quotes, four lines, 15 minutes.

And they're going to connect you with one of their expert brokers for unmatched service that goes beyond your policy. We use it at launch. It's easy, peasy, lemon squeasy. It's easy, breezy. What more do I need to tell you?

I use it. I love it. A lot of our startups, it, they love it. Try a broker today with the code twist and you'll get 10% off their startup package in broker.com slash twist. That's eM-B-R-O-K-E-R dot com slash twist

and use the code twist for 10% off. Okay, let's get back to this amazing episode. Give us another example of a tool. All right, that's good tools. These are like, this is like the content creation suite. You got SEO, you got clips. What's next? Well, what's next?

This is actually a request from one of the, listeners on Twitter. And he wanted to see how the new feature, which is called custom instructions works. Oh, okay. Inside a Chad GPT. So I don't know if you've played with this yet, J-Cal. I have not. So you have to go enable it in your settings.

Oh my God, there's settings now? I didn't even know there was set. Yeah, yeah. How did they add a settings to a GPD work? I mean, they really put out a really basic product. Didn't even have settings, right? Yeah, I mean, but, you know, look, they're doing it startup style. So I give them credit for that.

So what you can do now is there's two boxes. And so the top box here, I'm just going to read it, says, what would you like chat GPT to know about you to provide better responses and things that you can tell it is where your base and who do you work for? And then it can use all this in the context. And then another instruction you can give it is how would you like chat GPT respond? And here you can say respond formally or longer, make you keep your responses short. So all the things that maybe you get tired of putting in everything.

every single prompt, right? And so I'll do something like respond like a pirate, right? And so if I do that and, you know, we'll just go here and say, give me a short summary about, let's call it back to the future. And what this will do is you'll see here now, it is giving a response written in, you know, I guess, piretese, because that's what I gave it in the custom instruction. And this is obviously a simplistic example to highlight, you know,

how you can better use this. But for folks that are heavily using chat GPT, and I know you are J-Cal, I know your team is, this is a real lifesaver because you can start to now tune and customize. And this is the next iteration of, you know, where these LLMs are going, is like, customize them for how you need them and how you want them to interact with your personal information and also the style of responsible. I just, I just nailed it.

This is hysterical. Pull yours up. Okay. So it says, like, you know, what you didn't know about you? Where are you based? What do you do for work? I didn't put that in. But it says, how would you like chat chip you do to respond?

And I put in, I like to see data in a table. I'd like to see data in those tables with hyperlinks. Right? And I just ask it because we're doing the All-In Summit. And I always like to have like the best ice cream purveyors at the event. And I said, what are some good ice cream stores in Los Angeles? It says, of course, here are some populous stores in Los Angeles, saw and straw, yada, yada, jennies, splendid, grom, scoops, etc. And what's really interesting about this is it put the links in.

Yeah. I always tell it to do that. Now, if I didn't have that in there, would it have done that? I don't know. So I'm going to take these custom instructions out, and I'm going to ask it again, what are some quality ice cream stores in Los Angeles? And let's see if it does a table or not. It did not do a table.

And it did not do links. This is a game changer, for sure a game changer. So I have to go back in and make sure I don't lose that. Because I want those settings in there. Wow. That is insane. It's just to start. Yeah, it's just a, and think about this, Jake, how like.

It should be telling us what to put in there. So right now it has a little tool tip. I think you have it on as well. And those are generic. Again, my guess is in less than six months, those tool tips will be mined and farmed from all your prompts. Yeah, pretty great. Awesome.

This is a very cool tool. All right. Well, we got a short show for you today, folks. Sunny, anything else we want to mention before we get going to go in here?

No, we're going to save the rest of them. We got some other cool ones. Maybe we got to, I mean, and then there's so much going on in the news. I got to get a new show going. We're having All-Star Summer.

All-Star Summer has been crazy here. Yes. Some real hits. Woof. Woof. I mean, what a roster of murderers row. Oh. I know we had Darmesh from HubSpot.

We had Ryan from Qualtrix in the Utah Jazz. We had the CEO of MongoDB. It's been crazy. What's making the flywheel kind of go to the next year? So we did a, we slowed down to speed up. The show is not going anywhere, right? We're doing it for 12 years. I'm going to do it until I die.

So I said, okay. Make a list the 100 people who've been just incredible guests, you know, people you want to have on over and over again, right? So people always ask to have Chris Soca back on, although Chris Sok is a little upset at me because of Twitter. You know, my friend put Twitter. I didn't buy Chris. And Keith Verboi. Everybody loves when Keith Verbeye comes on. So you got those. You know, let's call those, like people who've been on the show three, four, five times. more. And people just love those folks. And then there's aspirational folks who we really want to have on. Like I've been trying to have Mike Ovitz on since his book came out. Yeah. It's just always

been like, you know, problem. You're just scheduling. And then, you know, we got other ones like, I always want to have Charles Schultz, the CEO of Starbucks. Starbucks. Yeah. So now we're making a list of the aspirational ones. And then, um, just like you might do a mail merge, uh, if you were a salesperson, you say, hey, these are my hundred targets. We started thinking about 200 targets at a time for the show. And then we start thinking about, hey, what would be a good time to invite them on? And then we do a custom email to them. So you showed that CRM system the other week that would be writing like the emails.

We don't like to do that. But if you make a list and you say, hey, for the next two years, let's work on these 200 guests. And let's really write them thoughtful emails where we talk about what we want to talk about, why we want to have them on the show, the benefit of coming on the show, etc. And then we send them links to people that we know they might know

who've been on the show. And we just can, you know, checkbox, checkbox, checkbox, you know, Glenn from Redfin, you know,

and then we're trying to get the CEO of Zillow for two years. We just, we got to stick with it. And so if you have a more systematic approach to the targets, that's what's driving it. Awesome.

Yeah. That's, and it's working, you know, as opposed to, hey, what should we do this week? You know,

it's like, let's really be, I call it like the term paper kind of approach as opposed to like the pop quiz approach. We can always pop quiz people like we had the guys from Varda on, you know, space station. They did the LK99. And then we're going to have the CEO of Varda on just to talk about Varda,

but then I was like, get him on now. You know, and then we'll still do his Varda episode. Yeah. And we'll just keep going from there. So that's what's really worked is like, you know, you can do the pop quiz. Hey, this is breaking news. Let's try to get them.

And then we can do the, what I call the term papers or the thesis. Like, let's try to get people who, you know, we've been trying for 10 years to get on the program, five years to get on the program. And that's a better combination. You know, a lot of people did like when I did the news with Molly on a regular basis. People got into like news every day. But it kind of alienated the core this week in startup's audience, which is founders who were like, I don't really care about the daily news or woke content or what Elon's doing at Twitter X.

and, you know, there's all these kind of like battles and tech policy. And, you know, if you want to, you have to really think about, you know, what is the purpose of the show and who's the audience for the show? The audience is startup founders and VCs. And they keep telling me, although they love when I talk about news, maybe not five days a week. Maybe one day a week of news, four days a week. I'm trying to get down to four days, be honest, because it's taking all my life. Four days a week, three days. One day news, three days.

other, you know, founders, you know, and then of the founders, all stars, legends, and then up-and-comers. Yeah. That's the combination. So that's what I'm really trying to balance here. I think it's like the Howard Stern show format, dude, you've got it down outside the skits, right? But you've got it down now.

We don't have, we don't do bits or skits. I wanted to do bits and skits. Really? I did. You used to do guess the fake startup. I wanted to get a person to come on who can do impersonation. and I wanted to have like fake Zuck on the show. And we did actually do it.

We had a puppeteer Lauren Feldman who would come on in the early episodes and, you know, we just tried to do it and try to see if we could have chemistry of like me talking to a puppet Zuck, you know, or AI Zuck. And so if they could, if somebody would make like a real-time AI Zuck where I could be doing the news roundtable and you could have a comedian who could say funny stuff as AI Zuck, that would be pretty hilarious. and, you know, just thinking creatively, like being able to talk to Steve Jobs, like some people made videos of me talking to Steve Jobs.

Yeah, yeah. Without being morbid or, you know, and I certainly would never want to offend somebody's family or something like that. But me being able to talk to people who are no longer with us be quite compelling in some ways, you know? And so I tried to do the bits. The problem is bits take time in production. And it's a different type of.

creative that you need for that. The one that we have been doing, I don't know if you saw, we did the Blackberry movie and lessons from the Blackberry movie with Lonhound. I didn't see that one. Oh,

yeah. Business breakdowns. One of our highest rate episodes of the past two months. So we're going to do business breakdowns once a month. It's kind of like Acquired FM does it or there's like,

we study billionaires. There's a couple of people who have this format of like, we're going to just take something you know, Nintendo, and we're going to talk about it for two hours. Yeah. So I don't want to crib that format too much.

I would call it a mix between acquired what they do and the rewatchables by the ring. Yes. It's a blend of the bowl. Yeah. So if you want to get in on that at some point with Lon. Yeah. What are we doing next, Nick? What's on the next?

I think we should do the founder next. Let's do the founder next. Absolutely. That's my favorite. The McDonald's one, right? Yeah. All right. Totally my favorite.

Also, because Mark Knopfler did the song boom like that after reading the McDonald's biography. So Mark Knopfler from Diochray, is my favorite. Wrote the song, Boom Like That, Because he had read Ray Crocs biography. Biography. A director heard the song, boom like that. And then got inspired to read the biography and then wanted to make the founder. And then in the film, they have the song, boom like that.

Can I also just say what I think when our show is the best, when it's the most helpful, when it's the most impactful, when it's the most different, right? Because that's really what we want to be. We want to be different because so much of this content is the same nowadays. I think our show is the best. And Son, you exemplify this in many ways. When the conversation feels like it's Jason and a founder or an investor or a peer at a bar at like 7.30 p.m. on a Friday night in Palo Alto. And they're talking about. Right.

Fly on the wall. Exactly. That is when I think the magic is made on this podcast. Which happened with Darmatian HubSpot. And Dave from Mombie B. And oh, he went. Yeah. Spelled by Dev, but pronounced Dave.

He went hard. He went hard on remote work and a lot of other things. He went. And then Ryan Smith was great. Tim Urban was great. Awesome. Mike Knop from Zapio was great. I like that new kid, Ben, from Soul Reader.

You see that Soul Reader yet? Have you seen Soul Reader yet? No, I haven't seen that. It's basically glasses you wear and it puts a Kindle in your glasses. So single-use device for basically not a VR, but you put glasses on. You're sitting bed at night. And you read and you just press the thing and next slide, next slide. And it's immersive reading.

Yeah, yeah. And I'm like, that's either going to be a total flop or there's going to be a group of people will become addicted to it. Now, I'm addicted to audio books. Like, I think this is going to be addicting for people to just be able to zone out and read without having to hold anything and just hit the next button. It's pretty neat. Very cool.

We'll check it out. All right. Next up, three great startups from Founder University. If you'd like to come to Founder University, founder. dot university. It's a 12-week program. We do it four times a year. Fifteen hundred people apply. Two hundred fifty people get in. Twelve-week course. We invest every time out of the 250 people who come. We probably invest in 20 or 30 of them and we give them 25K each for 2.5% of their company as they're like rich uncle, rich on, first money in just to give you

25K to get incorporated. If you don't need it, that's fine. We could do a direct investment if you're raising around or you can come to our launch accelerator if you get your product to, you know, a couple of customers, and that's 100K check for 6%. Founder. University, we're looking for a space, Sunny. So we're going to have an office, a garage. Wow. We're going to have founding university taking place live in San Mateo soon. So that's going to be a game change.

I'm going back to the office like you. It's way to do it. Been doing it for a year and a half, all right? I mean, I see NBC just asked me to come on today to talk about my tweet when I was like, listen, you've got to get back to an office here. It's like people are getting weird. All right. enjoy and we'll see you next time next time next time next week let me tell you about house of macadamias

this brand's got a special place in my heart the founders brandon and carman they are avid listeners to this podcast this week in startups and they actually started their company after listening to this pot and reading my book angel it's incredible it warms my heart and in fact their first angel investment was a huge winner and they used the returns from that investment to start a macadamia business so here are some amazing amazing facts about my favorite nut, macadamias. They have one of the best fat profiles of any nut, like an avocado on steroids. And macadamas have also been linked to fat loss and longevity. Why? Because they are the only nut rich in rare omega-7 fatty acids, which can support

natural collagen production and glucose metabolism and reduced inflammation. House ofacanamia products have no sugar added and no artificial ingredients. So many great ones. I like the dark chocolate myself to have the macadamacadamia bars. House of Macadamia never runs discounts, but they always give a special offer just for our Twist listeners. Get a free box of Namibian, C-Salted macadamacadamian nuts at House ofmacadamacadamia's.com slash twist. Plus, she'll get 20% off your order with code Twist 20. That's a $35 value plus 20% off with the code Twist 20.

All right, everybody, welcome back to this week in startups. we have a fun time for you today because our program, Founder University, has produced a number of interesting startups over the last cohort. And I feel like investing some money in these startups. I like giving them 25K checks as their first like Rich Aunt or Rich Uncle might do. The friends and family around. Kelly Shricker is with me. Again, she runs the Founder University program for us here at launch,

which is the name of our investment firm. Kelly, tell it to everybody who maybe hasn't heard what this pre-accelerator Founder University is all about and how it works mechanically. Absolutely. So Founder University is a 12-week program. We run it primarily online, although we do an in-person kickoff in San Francisco.

But most attendees will participate entirely remote. And from there, we basically help founders go from zero to one. So sometimes companies already have an MVP, sometimes they're just getting started with an idea. They build it during the program. Hopefully they'll launch, get early customers, and then some will also go on to raise some money. Yeah, fantastic.

And so the reason we started this was because we wanted to meet founders earlier in their journey, get to know them, watch their progress, so that eventually we could invest in them if they're going to build a company. That is venture scale. Not all companies are, and that's okay. Venture scale means you can get to $100 million in revenue, something in that range. So venture capitalists would be interested in investing. We do four cohorts a year.

We try to have 250 people in each cohort. Sometimes we blow past that. Sometimes we have a little bit less. We don't try to force it. Over 1,000 people, maybe 1,500 people apply to each cohort. We accept 250 at the max, although we have broken that sometimes. And then we try to invest in 10, 20, 30, 40 of those companies. What do we look for in companies at this early stage?

Half the companies coming into the program aren't even incorporated yet. So what we're looking for is teams of two or three with a developer on the team, you know, basically builders. If you have builders on your team, you don't only need a lot of money to get the product up and running. You can basically use sweat equity. So two or three founders is ideal. One or two of them being developers, two out of the three or three out of the three being actual builders. So designer, product manager, developer being the top titles or roles that you would typically see in an early

stage company. So we're starting our six cohort on August 11th. And we'll have, like I said, 250 or so builders in that group. You can apply at founder.com. University slash apply. There's still time to apply if you hear this next week or if you hear this the week of, we hear this before August 11th. There's still time to apply. But we do it quarterly. So not cohort six, there'll be cohort seven. And we invest in, like I said, I don't know, 10, 20% of the graduates. And sometimes we do it on the way in. Sometimes we do it in the middle of the program.

Sometimes we do it at the end. How many companies have we invested in so far? So far. In the last two cohorts, we're probably at about 22, all in with all five previous cohorts. We've done 41 investments and 11 additional companies have gone on to our accelerator. Amazing. Okay.

So here's what we're going to do today. We're going to hear some pitches from the last cohort. And we'll give one of them 25K. So we'll see three pitches. They'll pitch for two minutes, three minutes. What are they planning on doing? Two minute pitches. Two minute pitches, perfect.

So, and then I'll ask them some questions, we have a little bit of back and forth, and then I'll pick one for the 20. Okay. All right. First up, we're going to have Quario. Rami, can you please come on and show us your screen? Absolutely. Hello, everybody.

Hey, Romney. How's it going? It's going well. All right. And remember, most people are listening. So to the extent you can describe what's going on the screen. all the better. Three to go. All right. So my name is Rami. I'm the co-founder of Curio.

Self-service data analytics is a really broken space. There's a ton of really smart people who still can't answer their own questions and need to rely on somebody else in their company. And that's probably about Curio. Curio is a personal data analyst for every employee. And that doesn't mean analyzing data. That also means combining all your different data sources into one source of truth and giving you a way to analyze it powered by AI, all in an extremely secure way. Here's some ways that it looks like you have a very intuitive platform on the left. You can make great dashboards, as you can see in the middle.

And we also have a lot of over 300 integrations, and you can see some examples on the right. For the team, you have myself. I have five years of experience at Amazon, where I worked on a lot of machine learning and AI, both at Amazon Arbotics and Alexa. My CEO, Javier, who was a phenomenal operator with also a strong startup background,

and our tenured CTO, VJ, who spent the last 17 years trying to tackle the low-code data problem. We're already having a live movie. We're actually already making revenue. So even though we started about 10 weeks ago, getting close to three months. We released the MVP in four weeks, started generating revenue two weeks ago

and already have another 12,000 an annual recurring revenue in LOIs that we have not captured yet. Our initial target market is going to be small businesses with less and private budget employees that don't afford a data analyst just yet because they're the ones who are still feeling this pain point the most

and don't have someone to rely on internally. We're going to get to 100 million air in eight years and we're going to do that by charging at the per seat level because it's an individual contributor problem. I would need about 9.2,000 companies at 30 seats each to hit that number. For our route to market, obviously we're going to have our target market.

And for a more direct ICB, it's going to be marketers. We're also going to have a very strong outbound sales approach. We're already trying to collate a ton of marketing managers and just purchasing power people. So we can start that outbound and really start getting those customers inside for research and for higher revenue. For our customers, we're the only ones in the industry right now that are doing everything from front to back, vertical integration, to both actually give you that data and one source of truth and let you analyze it. Thank you. I hope you enjoyed the hitch.

All right. So fantastic. Data analysis is something that and having a data scientist that's only available to companies typically with hundreds of employees, tens of millions in revenue. For smaller mid-sized companies, it's an expense that they probably can't take on. But with AI, it's getting easier. And a lot of people have pools, lakes of data, whether it's Google Analytics or their stripe data, just databases. Generally, Google sheets. There's a lot more data available for you to plug into. So 10 or 20 years ago, people weren't storing their data. 10 years ago, you know, obviously three years ago, chat GPT and some of these data analysis tools weren't even available. So your timing is great. Cloud computing, Zapier allows you to connect things together. I'm assuming you're using Zapier to do all that. But they do.

We're not getting Zapier. How did you get all that, all those 300 in there? Is there some third-party tool that get you all those integrations? There was a third-party tool called 5chan. So they help us pipeline, and we're using an AI to actually do the transforming and joining. We were trying to do them kind of initially by hand, but especially for our go-to-market speed and to really get all the use cases immediately, we're using the third-party for now. Yeah, which is perfect.

I like to hear that because that's not your core business, getting the data in. So that's great. And then you look at the team, you know, a couple of years at Amazon for you as a product engineer. and your co-founders, front-end dev, and a senior software engineer, love that, and a senior product manager. So fantastic that you've got three builders on your team. Explain to the audience who's listening

what that means in terms of your product velocity. I think it means a few things. I mean, again, in like four weeks we got our MVP out, and I think that means is that we really understand the problem because we felt it firsthand in our previous roles, and we just also have everything in-house to make it, built really quickly in a very scalable way. And I think one thing that we've had a lot of early customers talk about to was just security.

And because of our CTO who has a lot of data platform and experience, we're going to have SOC1, SOC2 compliance right out of the gates, and we're going to be able to actually provide that assurance of security and privacy to the companies because we are taking the, you know, we're working with their most intimate information, right? It's all their personal data. It's all their company data. They want to make sure that it's somewhere safe.

Yeah, fantastic. So the market is a great market. You got a great team. You got product velocity. You've got a tested business model, consumer subscription, I'm sorry, business subscriptions, SaaS products tend to do great. And you got some early traction. So you check off all the major boxes that a seed investor like myself would like to see

and a firm like ours would like to see. So congratulations on that. Next up is before sunset. Batu. Am I pronouncing your name correctly? Batu. Yes. Yes.

Great. Awesome. Two minutes on the client. three to go. Hi, I'm Botswan, co-founder of Before Sunset AI, where we help people plan and execute their day with ease. Meet Paula, our ICP. Paula works for a hybrid team working on multiple tasks and projects at once. Think of an analyst in an investment firm or a software developer, for instance.

In an ordinary day, she finds herself jumping from task to task, app to app, and still doesn't feel like she has accomplished much or has visibility to team and project status. Introducing BeforeSonset AI. A platform where you can maintain to-dos, have AI plan today, view your calendars, see how your team is, and what they're working on. It can also process your data in the analytics page to help you improve how you work. So, we've been busy in the last couple of weeks. We launched our product and product on and finished number one product of the day, week and month, onboarded over 10,000 users and kept people coming to our site to convert following the launch hype.

Our continuous revenue streams include individual subscriptions and enterprise plan, charging monthly or annually. The competitive landscape has various players, but we distinguish ourselves with three main capabilities, AI, personal analytics, and team features that are already built into the app. Like I said, we launched a little over a month ago, and since then, we've been putting in the effort to reach out to our ICPs. These efforts include smart SEO tactics to keep growing organically, small influencer campaigns, and some pilot ads. The calculated market size is $5.8 billion, taking into account 28 million potential users and average $15 monthly fee.

To get the 10 million ARR, we need to maintain about 50,000 paying users which we aim to get by 2026. So we're three co-founders, and each of us have our own unique story about how we found the passion to create Before Sunset AI. Again, I'm Botsohn, co-founder of Before Sunset AI, helping people plan and execute their day with ease. Thank you for your time. Okay, fantastic. So using AI, pulling in all of my to-do list, pulling in my calendar, and then telling me, hey, this is what you should do next and maybe doing that with my entire organization. That is the goal here. Correct. Yes. And so is it a,

do you see it as a scheduling to-do list or project management like Asana, to-do list like you can do a notion or Coda or you know your native to-do list in your preferred operating system, Google and Apple have that and then scheduling. Obviously some people

schedules equal their to-do list. So where do you see this sitting? Is it competing against Asana and those other tools? Isn't something new? No, it is a different approach for sure. Our main purpose is to give a tool for the users to be better to be able to better

plan their day individually and then kind of scale that to the team in the long run to show them what they're working on, what projects they're working on based on the tags that they set and how much time they have on their day before they log off and stuff like that. So do you see this being something where you're doing time tracking like law firms or consulting firms are saying, hey, we're going to do time tracking. From 9 a.m. to 9.50, I'm going to be working on this billable project for this customer, or just, you know, put my time, if it was Kelly, this is my time for Founder Universe, is my time for some other project we're working on, the podcast, etc.

Or do you see it as monitoring? Because now there's like screen time on your phone, where it monitors and then gives you feedback, a little bit scary. There's obviously screen monitoring, stuff like that people use for remote work, but there's also monitoring systems that are more for optimization. Hey, you spent, you know, two hours on Reddit. Oh my God, you're wasting time. Unless, of course, you're doing marketing on Reddit, in which case, that's like a positive thing, right? And so people are trying to figure out some, I'm interested in the modality of how people record their data or if you're going to be recording it with a system tray item or they're going to explicitly fill it in. How do you see that data being acquired into the product collected?

So the comparison, especially the comparison between that monitoring versus like your individual push to record the data on what you're trying to do and what you want to do, we're definitely on the latter side because we are trying to provide users a more mindful productivity tool rather than them being always kind of looking over their virtual shoulder to see who's keeping track of what they're doing. Time tracking is definitely a key part of what we're trying to do. It's a key feature that we've added in, so the integration into the calendar can be done more effectively.

And the insights that we provide them in the analytics page once they recorded can easily be translated saying you thought you would spend like 30 hours on a specific project this week, but you ended up spending 40. If you're, you know, setting yourself to be paid for 30 hours on that project, you just, you know, undersold yourself. Just giving the tools to the user to make sure that they're earning their worth, they're not burning out, and being their best self and best worker.

Yeah, so I think there's a lot here. I love the fact that you've got this great team. and with remote workers and people wanting to optimize and the mistrust between that's kind of brewing and the standoff go back to the office, work from home, tracking time, who's screwing around, who's overworking, who's working two jobs at the same time. I mean, there's so much chaos in the space that I think any tool that can make management and individual contributors feel like they're more in sync. we share our calendars at our investment firm. People do a start of day, end of day, and they do a stand-up. So the total ask, I think, is five minutes at the beginning of the end of the day to sort of plan your day and recap your day, maybe round it up to 20 minutes if people put a lot of time into it 10 and 10.

And then do a stand-up every day. It's a half hour with your team, 20 minutes with your team, just, hey, here's what we're going to try to accomplish as a company in the next 24 hours, next week, next month, etc. So I think that's, this is a key, key opportunity for you. And then you just have to pick your, who's the ideal customer here? Is it small teams? Is it big teams? Small teams tend to embrace this stuff. High performers tend to embrace this stuff.

And then people who have to track their time, they're also looking for solutions. So people who are doing billable stuff. So I think you got like two or three different ways to go after these users. Who do you think you're going to go after? the time, people who track time for billing or people who are just super optimizers, quantified self, people, productivity, get done, GSD people. So, uh, ideal customer being. So definitely, uh, the GSD population is a little bit more appealing with respect to the

fact that the people that bill hourly, like lawyers and such, they do have very specific tools that are already available to them. Yeah. Before sunset is a tool that they can use as well because those tools don't have. AI. So if they want to lift some weight off their shoulders, they're more than welcome to, and it would be a great addition to their tool stack. But in reality, these productivity hackers, the GSD folks that want to make sure that they optimize what they're doing and how they're doing it would be our prime candidate. And you want to go solo, single player to start for the first

year or two and then maybe add team features in year two or three. Is that the goal? The launch of the individual aspect was June 15th of this year, and all the users that we've accumulated at 10,000 plus have been on that field. Now we're planning on launching the team side of it this fall, and we're going to be rolling out the beta version, onboarding people, some small teams to see how they like it, continuously collect feedback, and make sure that we can scale it to larger teams or a great number of more number of teams.

Great. And so for those of you who are watching the show, we're listening right now, the interaction you're seeing here, what I'm trying to do as an angel and seed investor, early stage investor, is to listen deeply to how the founder thinks about their business. I'm giving them how I think about it. And what I'm looking for is, you know, how certain are they or how deep are they into it? So I'm presenting different options here, Kelly. So you think you're going to go after the time tracking people? You think you're going to go after the productivity people? What do you think about this industry trend, remote work, going back to the office? I just, I don't even actually care about the answer as much as I care about hearing the founder's thought process and their perspective. Is it a considered perspective? And we see here is you've really thought this through, haven't you? Yeah, definitely. So, and the passion also comes through, Kelly, when you have these conversations, right? You can start to identify, hey, this person really does care about this problem, this product, this set of users, or, you know, a,

the wider word market space, right? Yeah, absolutely. We talk a lot in Founder University as to what is the plan. And then how do you talk about the plan and get people bought into the plan that you think is ahead? Yeah. And people who make plans tend to perform better,

people without plans. Kelly and I have been working on the plans for Founder University. We have an ongoing dialogue. Hey, we want to have 250 people, 1,000 people go through the program a year, four cohorts, how many applications do we need? Do we need 1,000? do we need 1,500?

What are the qualities of the people coming to the program? How do we communicate with them after they graduate? How do we invest money in the winners? You know, if you set some goals, you can have a really good dialogue of, well, maybe 250s, too many people. Maybe it's not enough. And, you know, this is all, I think, good things for angel investors who are listening to the program.

I know a lot of you listen, syndicate members, other venture firms. It's really good to hear about the plan. Okay. We got one more to go here. And then I'm going to announce which one gets to 25K. Kelly, are you ready? Tell me about this next company. All right.

Alan is up with Symantec. Okay, Alan. Tell us about your coming. Okay. So, welcome to an overview of Semantic, the AI co-pilot for product design. Semantic works almost like magic. Just describe a natural language, any kind of application you would like to have, and semantic creates the user stories, the user flows, and the entire user interface for you.

Oh, instead of spinning. hours and days building a user interface by hand, Semantics AI can generate your product interface instantly. You can then test with a clickable prototype to get a feel for the overall experience and view your design in different branded styles by simply swapping design systems on the fly. Whenever you want, you can export your screens as layered files for Figma or as fronting code into any repository. It's an AI co-pilot designed to fit into any existing tool chain and workflow. We've started to onboard paying customers starting July 4th, and we have over 300 on our wait list. And additionally, we have verbal commitments for 10 enterprise licenses. We have a freemium

B2B SaaS business model with per editor pricing starting at $45 a month for individuals, $85 a month for teams, and $120 a month for enterprise. We're building semantic based on years of working in design and product teams, building and procuring product tool chains, and understanding what the real problems are. We believe we're very well positioned to differentiate ourselves effectively with our focus on integrating into the existing enterprise tool chain and workflow. Today, Symantic is a web application with a Figma plugin that enables designers to export their assets. We're building on this to make semantic an integrated necessity in any digital product tool chain and workflow with features like multiplayer mode, an extensible plugin system,

and an API, which we believe are key to building a large community of users. We launched our technical alpha at the beginning of May, and we launched our paid customer beta the first week of July. We're on track for our commercial launch in Q3. Our average customer is worth $750 a year and gives us a very realistic path to our first $10 million in ARR, and then on to $100 million. We're a team of two full-time founders

with strong backgrounds in design and development in enterprise, agencies, and startups, that's semantic and AI co-pilot for product design. Thank you. All right. Fantastic. The demo is amazing. I don't know how much of it is reality yet or how much of it's been realized, but it's pretty clear. I can send you, I can send you an invite so you can try it out. Correct. So what we see on the screen exists in the real world. Yes. Okay. And so this is something we always like to just get a handle on. It's okay for founders to show us what's coming in the future, future product designs and in a demo, we sometimes have to just

clarify that, but this is a really great progress. It's a great vision. Everybody's talking about this. Can I just talk to my computer with AI and have it understand what I'm looking for, have it build the designs for me, eventually build the code for me, eventually publish to the store, and this is why startups, instead of needing 10 people to be, you know, a basic app company, it's going to go down to six or three. Because tools like this are going to make people bionic. We're already seeing it with co-pilots for writing. We're seeing it for co-pilots for writing code, co-pilots for doing images and design. And of course, why would U.S. be any different? It's going to make people bionic faster, get more looks at something, just like templates really help people build better webpages. So I love

it. I think the pricing is good. You know, if you can get people $500 a year, $1,000 a year, That's the right price for something like this. Maybe not 300 a month, not three or four thousand a year, but designers, if they get some value from this, it's no problem going to their boss and saying, can I spend 500 bucks or a thousand bucks a year on something? Maybe they do that already with the creative cloud and Figma, right? So really, really interesting.

And yeah, it looks like you and your co-founder have a lot of experience in UX. And competition, I guess, is the big question. I see the competitors you listed were a lot of the startups building AI UX companies and prototyping companies. Do you see them as the real competitors or you see eventually the Figma's envisions Adobe Creative Cloud as the competitors? So I think it is really going to be the startups. This is an AI native approach. It's totally rethinking how we build digital products. and if you look at the incumbents,

you know, they're heavily invested in these very professionally oriented tools. So, for example, Figma, the last config, you know, what they announced was they're sprinkling some AI into their existing tools to reduce a little bit of friction, but they're not going to rethink the whole product. Got it.

So it's still going to be a professional designer oriented product. Got it. And so who do you think the market for this product is? Do you think professionals who want to dip down here and use this as like a scratch pad? Or do you think it's more, hey, some entrepreneur maybe doesn't have a UX background. They don't want to do Figma or envision. They don't want to be sitting there on the grid learning that tool. They just want to brainstorm and sort of build stuff.

How do you think about who the actual customers? Are they going to use both platforms, Figma and this? Or is this for a new category of user? right now I would say we're kind of a sidecar to these existing tools. We're not trying to replace any of them. Are sort of the person, the profile that we're going there is someone who's building digital products. And the challenge or the problem that we're solving is that traditionally you have a cross-functional team of a product manager, marketing people, developers, designers. And they all have great ideas, but then they hit a choke point because there's usually only one individual who can visualize those ideas.

Right. And give people, you know, what this is actually going to be. Just like the data scientist in the company. Exactly. Exactly. Everybody needs to get data. They've got this big punch list. Who am I supposed to do the CEO?

The VP of sales, the CTO. Everybody's got a different pet project. And here you could, you know, be somebody in sales. or somebody could work in marketing or somebody could work in the C-suite and they could be making designs if they want to. Yeah. It's great.

Enables a whole different class of people, just like Canva did, right? Like people could go in Photoshop and build stuff, but that's a lot of work to learn Photoshop and doing the filters on Instagram or building stuff in Canvas is just easier. So it makes a lot of sense. Really, really enjoyed all three companies here, Kelly.

So when we look at them, what you see is really motivated teams, two or three co-founders in each one, and all of them are going after some sort of AI angle, right? Using AI to go after a specific niche. Now people are going to say, oh, well, can't chat CPT just do this eventually? You just talk to Open AI or BART and we'll be able to do everything. My belief is the answer is no. I think they're going to be verticalized specific tools that build all the accruity.

around this and they'll be optimizing at a level that the base platforms are not going to be designed to do. So that's my personal belief. I think when we had Brian Chesky on from Airbnb, he had a similar belief. So really difficult decision here. Which one of these do you think will be the biggest return? If you had to pick one, would have the biggest return for our LPs, right? We're investing other people's money and mine to hopefully get a return. So you had to pick which one, Kelly, do you think? I'd ask you which one has the best product velocity and then which one do you think has the biggest ability to become the largest company with the most revenue therefore to get the biggest return. Absolutely.

So semantic is my choice for biggest return. I do think that any time you can empower team members to do work that needs to get done, bring things to life outside, just like Canva has, that there's a big opportunity there and companies are going to be willing to pay for it. In terms of product velocity, I was really impressed with before Sunset. When they launched, they had over 2,000 users, which tells me that there is a need and a want in this space. It's definitely something that exists in a lot of different ways. It will be interesting to see how they make it really a tool that you have to go back to every day and stick with and how it can grow virally. But I do think they're on to something. All right. Fantastic.

So now I have to make the hard decision of which of these companies to put 25K into. before sunset, AI productivity tool to organize and prioritize your schedule and your to do list, your tasks, great idea. And Quirio, I'm pronouncing you correct, yeah, Quirio, and AI data analysts, what could be better than that? Josh, that's just such a compelling vision. And then finally, Samantha, God, I'm a product guy and a tool to help. people build products faster and better and make everybody bionic is just, that's the holy growl of startups right there. So bring the three founders here, Batu, Rami, and Alan. Difficult. You put me in a

very difficult position here. To pick only one of these companies would be criminal because there's so much potential. But that was the promise that we made at the beginning of the program that we'd invest in one of them. But you know what? It's my money and I'm the GP of the fund. I can change my mind. I want to invest in two. I could do that. I could potentially invest in two. But you know what? I'm not going to do that. My gut, Kelly, tells me, investing in two is not a good idea. I'm going to invest in all three. I love all three of them. I loved all three of the pitches. I loved all three of the teams. So we'll send each a 25K check. Hopefully that helps you get your legal bill started, get you incorporated. And then of course, you know, we have our accelerator and we invest larger amounts

of money as things move on. So consider this the starting point of the relationship. We like to try to get to 10% ownership and the winners of the companies that we incubate or have at our pre-accelerator, found university, launch accelerator. So congratulations, everybody. It's good to be in business with you. I'll see you over in our founder Slack. You now are in the founder Slack. We've now been on this week in startups. These are some of the advantages of being a launch company. get to meet a lot of other founders and hopefully we can be supportive and helpful to you as you grow the company and of course introductions and meeting other investors in the valley here. So it's good to be in business with all three of you. And Kelly's got some bricks of cash.

I can't see them not on in front of the screen, but she'll just drop some bricks over. And yeah, welcome to the team as it were. Welcome to the club. Congratulations, everyone. You all have worked so hard. Thank you very much. Jason really appreciate it. Thank you.

Yeah. Fantastic. And then maybe just as we wrap here, a little bit about, hey, the program. What works really well in the program? What do you think we need to improve on for Founder University? If I can jump in. Yeah. First, you have a great team.

So impressed. with Kelly, with Prash, with everyone I interacted with. They're doing an acceptable job at this point. We don't like to give too much praise here. It's an acceptable amount of effort. I would agree. I concur. You have an embarrassment of riches, Jason.

The content was good. It was well structured. Oh, good. Yeah, we've been working on that curriculum. Shout out to Charlie, who helped us build the curriculum. He was here. It really helped connect the dots. So maybe a lot of this stuff I'd sort of picked up.

here and there, but this really drew a line through it. That's good for us to be. Yeah. Yeah. And Kelly's coaching on my pitch was just really invaluable. So thank you a lot for that. Fantastic. Fantastic.

Remy or Batu, any thoughts on the program that you would like to share. Yeah, I'll quickly, you know, pitch in here. I think it was, Alan did mention about the curriculum. them. I think the timetable of the curriculum was also very accurate. We had the products just getting started as we started Founder University and some of the information on how to utilize maybe some no-code applications if we needed it or some marketing tricks that could utilize. All of those were taken by heart and helped us get to where we needed to. So they were greatly appreciated. And I I love that there's a program like this to help all the founders that are out there,

250 at a time, to benefit from it. You know, like I think we hear this a lot, Kelly, hey, I knew a lot of what's in the curriculum. I knew 60% of it. I knew 70% of it. I knew 50% of it, whatever it is. But what you don't know, you don't know. That could either be a huge opportunity or it could be something that could scuttle your startup.

So we've really tried to look at, and this is one of the benefits of having over 10 years of investing in portfolio. We know what's working in the field. It does change. Some things are timeless. Other things are not timeless. You know, certain tips, trick, strategies, they might change. Strategies might not change every 10 years, but tactics might change every year or every two or three years. You know, some TikTok or Google search or product hunt as a tactic.

All these tactics could change over time, which, you know, how to use. these things. And so it's really great that we have this organic living document. What we learn with our breakout companies that are going public or, you know, are reaching a billion dollar valuation or raising $100 million. That informs the curriculum that we've put together. And so as a but a couple of examples, just making sure people have their accounting, you know, dialed in, making sure people know how to pitch their company and understand how venture capitalists and see investors make decisions. These things are really important for founders to know. And not, So much of the advice, you know, I find out there is low quality.

What we've really tried to do is find, you know, from our first principles and our experience. And then also what other really successful people are saying. And we get to have a lot of them here on This Weekend Startup so we can point you to a This Weekend Startups episode or, you know, a Techstars blog post or a Y Combinator video. You know, we don't have an exclusive on knowledge. We like to point to the best information we can find and that founders tell us has been the most. effective for them. So, fantastic. All right.

And I think we'll make that a wrap. Nicely done, Kelly. And we'll be doing some live events. Did you guys get to come to any live events yet or no? No. Okay. So I owe you a live event. We're going to be having more of these. I'm going to try to do things with the alumni in New York and in San Francisco

on a regular basis in San Mateo. So we'll look for more in-person events. We'll always have the remote option. people like to do the remote thing. A lot of people are, you have maybe have jobs and we do it at night. So it's an hour and a half Monday and Thursday nights. And so, you know, you can still work your day job while you come to Founding University. When you go to launch accelerator, you should probably be full time by that time. So we have a tolerance for people who are moonlighting and doing this as a side hustle. And hopefully we can help be that bridge with that first 25K check. Go to founder.com.

Find out more about the program. Kelly, thank you so much. You're working really hard, doing a great job. where I should say, you impressed you're doing an acceptable job. It seems like you've checked the box of acceptable performance. So, congratulations. I don't know. Really great job. I'm really pleased with the quality of the companies. We're really meeting great founders early.

So that makes me very happy. I'm very impressed with them as well. And getting to know them during the program, especially this group here, is amazing. You know, they're really working hard. And it's a pleasure and a privilege for us to be able to be your first investors. where you like to be the first investors in these companies. We'll see you all next time. Bye-bye.