The 19-year-old Who Built a $1.5M Ai SaaS in 7 Days

0:00how much money do you make so in our first year business we're going to make over like around 1.2 to 1.5 million million dollars yeah how much how old are you I'm 19 wait I didn't even know that you you're making $1.5 million this year and you're 19 yeah okay what do you do so we built software products so the one one of the biggest products we have built is if you've seen on Tik Tok you've seen all these AI Drake covers and all these AI SpongeBob videos so we actually built the tech for that and uh we launched it about eight nine months ago and very

0:30quickly went from zero users to now almost 2 million users wait like that Viral AI Drake song yeah so that was one of our early early users that so we found this product like it was essentially like voice too conversion we're like there's so many use cases we allowed a couple of beta users about like 30 40 people one of them actually made that song Put it on to the internet and whoa that's really cool just to see someone that like used your software and like went hella viral from that essentially yeah it was like all of our distribution done for us what's it called it's called music file music F

1:01and it's still like running right now yeah we we still have like around 1 million active monthly users which is pretty cool wow holy so as you've gone about launching this cuz how long did it take you to make this so it took me a week this is the story yeah so what happened was um over the last year I've spent working at a company called buil space which uh so I was 17 18 and I was working at this company called build space they're essentially building a school for people to follow their dreams rather than the traditional pattern of going to a tradition school and like

1:31following a fixed path imagine you got to work on what you actually wanted to work on whether that's be a musician or uh make it make an AI product or become an athlete whatever it is like you want to go do something how can we help you so we launched a six we program called build space uh which was nights and weekends and essentially for six weeks we put you in a group with other passionate people about their own ideas and essentially in those six weeks we help you go from Step Zero to step one which is essentially starting starting in your idea so during that I learned a

2:00lot from like the the founder his name was farza I was really young 16 17 18 and I spent a lot of time just learning how you would scale a startup they ended up raising $12 million and and I spent a lot of time build building the startup in that time I realized okay cool to build huge companies you're going to have to have huge missions and huge goals and I want I needed to become an engineer as well I needed to learn how to write code because everything I did required code back then and I was just an engineer but how do I build my own

2:26products you also have to know how to write really good code so after I left I decided I'm going to spend the rest of the year kind of just building projects every month so every month I just built a new project so I was just I was trying to refine my coding ability and then one month I came across this research paper which was talking about the the vocal style transfer that a research paper in China released and I realized like this can be used for music and we we saw like a couple of random like AI Kanye songs and this was before the Drake thing

2:53there was the first thing was AI Kanye and I was like oh this is really cool I wonder how people do it uh and I came down this Rabbit Hole of realizing izing that people were in this Discord called aiub which was like a really small Discord at the time only eight or five or 8 thousand people and I I remember in in the chats people were like hey what is a print error and a print error is like the simplest type of error in coding and I was like these are like non-developer people trying to learn how to convert their voice to Drake or trying to learn how to convert their

3:20voice to kany and I realized like if I just built a simple site with the tech that's already out that people are going to want to use it anyways because they're trying really hard so I spent a week from that idea and kind of just launched it and a week later it had 100,000 users so 100,000 yeah so from idea to launch 7 days and then from launch to 100,000 was seven days was it free for the 100,000 yeah it was free and I was losing money so I didn't I didn't monetize it at all I kind of just released it just thinking like it would

3:45be cool God and uh people really thought it was cool so there was was there any productized version of that like of AI voice covers no so we launched the first thing and that's why I grew so quickly and also at the same time we didn't really think about we didn't know it was going to be like a company we kind of just watched it as a side project thought some people would use it didn't think it would have high retention and very quickly we realized it was the opposite people really liked it people were sharing it and it was super simple it was like select between Drake Ariana

4:12Grande and Kanye West and you just upload a vocal file and press convert that was the entire site but we were spending about $3,000 in GPU credits every single day for those first weeks how did you afford that I didn't so that's that's like the craziest part so I I launched it for fun and I was like okay cool maybe I'll spend like 5K on it and I'll have it on resume is like a cool big project I built and then very quickly I realized okay like this has gotten to like 10 12K I should probably make some back so rather than like thinking of a really proper monetization

4:40strategy I was like okay I'm just going to put like a $1 like pay wall like $1 subscription per month and you can use the tool in the first hour of me launching that I remember we had like $95 of $1 subscriptions paid and by the day end of day one it was 2,000 I was like yo so 2,000 people paid a dollar today so I feel like I could probably raise it a little bit bit more um and so that's what we did sass in 2024 is all about getting your idea to Market as quickly as possible that's why we love no code SAS we can use tools like bubble

5:11and make.com to make an MVP in less than a week and then validate it by bringing it to the market so no code tools make it possible for anyone to build software now exactly how Shopify made it possible for anyone to start an online store bubble andm make.com make it possible for anyone to build web apps so we have a brand new program called the future Dev where we teach you how to build virtually any software idea you have all with no code so if you want to learn how to get into the SAS game and build your own software click the link below and join the wgmi academy today how long has

5:44it been active so we launched in April this year wow yeah not too long ago eight years yeah eight months eight months yeah sorry when you were 11 okay so it's been around for eight months and you're already at over 100k a month in Mr how did you like get this out into the world like how did you Market it so we realized early on that to create any type of viral product you want to make a lot of controversy that's essentially get like both the people that like you talking about you and the people that don't like you talking about you so when you first launched that AI Drake and

6:16weekend song it already stirred up a huge fire like people were like oh this is going to be great for the music world and some people were like yo this is terrible as an artist and musician um so that was like good because people are now like coming to this understand like this is a new thing the market then we ended up basically spending a lot of time with ugc content creators and essentially created like this crazy affiliate program if you people were already posting AI Drake videos and AI SpongeBob and all that already so we were like hey how can we provide them an incentive to post about our platform so

6:46we said hey here's an affiliate link you're already going to post this content might as well make some money while doing it so they started driving traffic to our site and getting paid for every sign up so we ended up having around 4,000 people make like Tik Tok pages and just started in our links out of there so that's kind of how they knew it was coming from your platform exactly so when that Drake song We viral did people know it was coming from your not originally at that point we hadn't launched like publicly it was still like just like our early beta beta users but

7:13like a week or two later we launched publicly and that's when people started to see like okay this is kind of how you can do it yourself if you wanted to that's really cool and so mainly just an affiliate program that what was the percentage like 50% or it was it was not even that high it was 25% but the the main thing that we did was like people are already posting so much AI generated content we just kind of went to these big accounts that had like million they're growing very quickly at the time too right because all over Tik Tok and you're like hey you're you're already

7:39going to post this content like we can help you make money while you do it so here's 25% some people are like 30 40% we can do custom custom Affiliates and we kind of just gave them the link and that's pretty much it I think to till this day like one of our biggest Affiliates has broughten in $50,000 wow yeah so it's like crazy and so he's just posting AI generated content and now he just threw your Link in the bio for people why not right and and his Tik Tok Pages really well and and that's it huh are there any other like like organic marketing hacks like or viral like Loops

8:09you set in there for people to like see yeah so there's there's two things I would say about like our distribution plan one was Tik Tok really early the problem with Tik Tok is you get a large amount of traffic but it's not high quality traffic and what that what I mean by that is like if you're scrolling on Tik Tok and you see an AI cover it's not very targeted right it's just like some random person could be just chilling in their home see this cover go to the site maybe try it but they're never going to purchase a plan yeah so we'll get millions of users or hundreds

8:35of thousands of views coming from Tik Tok not really converting but then we started doing long form content as well we had we had a couple of influencers that made like YouTube videos about us like here are some of the coolest AI music tools one video had 300,000 views and that brought in $30,000 in Revenue in a month compared to like our 300 million views in a month and that only brought in like 20 or 30,000 so there was huge contrast the density of YouTube yeah what was the person that posted that like a producer or like yeah was it was a producer like a music produc like

9:02teaching people how to do it yeah so essentially what we did at the beginning we launched as this tool to mess around kind of convert your voice into like Drake and Kanye and then we got a C AIS letter people actually don't know this but three weeks in we had a letter from Universal Music Sony music and Warner Music I have it framed in my wall it's literally like it's actually like pretty crazy to get a letter from all three of them at at the same time and essentially what it said was hey you can offer Drake or Kanye or all these big artists on your site because you're essentially

9:29you're T you're selling a product with their branding and you're making money off of it so we're like okay we agree with this we won't so that same day we deleted all these Dr Kanye models all these big models and replaced it with the ability for people to train custom models that's where our pricing came in so now people if they want to convert their voice into these ex artists they kind of have to train custom models themselves and they have to pay for it so that's kind of how we got around that so it's taking the accountability off your guys' shoulders in a way so instead

9:55of offering it Drake and Kanye they have to come in and if they want to you contr it essentially we made it focus more towards like the music use case so we essentially what we did was we replaced all the the the big famous artists with 50 artists that no one knows essentially we took 50 people's voices and we took like 50 other people's voices so 100 100 voices took 10 minute recordings of each voice so there's 100 custom voices and essentially what we did was this thing called linear interpolation where you take the model weights of 50% of one model and the model weights of 50% of

10:27another model and mix them together and you said create a new voice that never existed you mix two people's voices and create one new got okay and so those voices actually don't have copyright like they actually don't exist so we essentially offer them on our site right now if you go to it uh and artists and producers can use them in their music freely rather than hiring vocalist uh for example one of the big examples that we use is a lot of times we have male singers that want female background vocals or vice versa instead of hiring a vocalist paying them royalties and doing all the studio time now they can just

10:55sing themselves and convert into any voice they want this is super empowering for for producers and Engineers that are like trying to like up their skill cuz like you just said like the hardest part is finding quality talent but if you have no experience or no money you can't afford someone's to like have someone use their vocals so now they can just literally get some of the highest quality vocals and then really show their creativity I'm on the side this going to be really positive for music yeah it's it's very 5050 what what we've realized early on uh so we have a lot of users now but early on we realized the

11:23people that were coming to us very quickly were like the A-list celebrities like we have a lot of ndas with really big players and the reason I thought that was because a lot of these big people the way they St stay ahead of everyone else in their Market is by adapting really quickly so a lot of big a A-list celebrities one one one that we have publicly is leis Bell leis Bell is postm producer he's like known as the best producer in the world so he's been using our prop platform for a while as well and what's he getting out of it the most so he's been using it to just

11:50essentially play with the music ideas convert his voice into other artist voices to send them Demos in their voice that makes so much sense so just a bunch of stuff like that yeah so you could like literally come with an idea for a song say hey Travis Scott would sound great on this and just do like a loose version of it yeah if you wanted right and then send it to him their team yeah wow that's awesome that's really cool that makes a lot of sense it's like a good practice tool wow good job man okay so this was build space is basically like this just after school program or

12:17how did you it's it's online incubator essentially it's an idea incubator if you have an idea whatever you want to do online yeah it's online six weeks online uh you kind of just spend six weeks online and there's online tutors that kind of give you Frameworks and like here's how to do distribute anything whether that's music your own product whatever it is here's how to think about uh consistence here's how to think about XYZ so it's just like a credibility format it kind of like Y combinator in a way it's it's like why combinator but rather than for startups it's for any idea and they went through y combinator

12:44as well oh cool did they you have to pay for it did you pay for it it's free so we believe the education should be free and uh so that was the big big goal behind starting and we launched it and it's doing really well it has like 200,000 students that I've gone through now wait are you a part of the build space I I worked there for like seven months oh you worked there yeah yeah so the founder his name is farza he's like 26 27 he it's actually his birthday today so he's the person that kind of brought me into this world of Entrepreneurship and gotcha founding

13:09yeah okay so you were like a part of the team of build space and you happened to build your own thing in the program as well I actually yeah so as soon as I left build space I was like okay I'm going to spend the next couple months kind of building it and then I was like I never even did my own like company's program so I ended up doing it and in that the first time I ever did it I was like built this thing do they take like a percentage of they take nothing so they're they're like they they don't have a business model in terms of right

13:32now they liter just don't make money right now right now yeah so they they just raise like a crazy amount of money and they're focusing on just figuring out how to incubate ideas but their business plan in the future is going to be around the idea of like getting subsidized education from government so right now they're partnering with Dubai like building schools there for entrepreneurship that's awesome yeah the government kind of pays for wow I want to talk about that that's really cool but let's let's talk about your plan for the future of your app so so you're at 100K a month you basically launched it

14:02really quickly blew up really quickly you figured out how to monetize it after getting ceas and assisted yeah so what's your plan as you go into the next year three years because you said you wanted to make a big company you a big idea so around music I've this is like I never thought I would be making a music company I've never even gone to a concert so I haven't spent a lot of time like in the music world yeah um but I've done a lot of research in around the market and the Tam and like the size of the market music tool softare is are

14:29like the lowest Tam like it's around like less than a billion dollar market like Pro Tools yeah exactly ablon like these these tools are now worth over a billion dollars they're all less and I want to build a a legit company I realized you can make a lot more money if you not only solve music problems but also audio problems across other Industries whether that's like audio in Games audio in animation movies and what I mean by that is like all these characters in games all these characters in movies have voice actors imagine you never have to hire voice voice actor again imagine one person can voice act

15:01an entire movie right so that's like something that we're working on we're also working on the ability for people to generate music for their games and for their movies because they also have to pay for huge amounts of soundtracks which cost tens of millions of dollars another big thing which I'm really excited to launch in the future which we already have a beta working is essentially in all these games and animations you have so many sound effects a man's running with like lightning in the background someone actually has to go in a studio and make that sound imagine you were able to just that via text text to llm and audio

15:31comes out is that exactly what prompt so you really focus on B2B right now exactly right right now we want to build this B2B product which is going to be like this for for gaming gaming companies because we have so much interest from them already we just had an investor reach out who just uh invested in our company Jer slav he's the founder of beat saer which is the biggest VR game in the world he sold to to um they sold to meta for $2 billion and he did yeah beer billion two billion with the be that's how like so music is a25 billion market and and gaming is a

16:03$250 billion Market which is expected to double over the next three years right it's 10 times bigger so there's a lot more money to grab and especially in the audio problems that there's not that much Innovation happening so yeah yeah audio Engineers are like especially sound effects and sound design for movies like that is like an art form in and of itself yeah it's like so I can imagine how hard it would be to come up with how to make these sounds because they literally physically have to make them in the studio yeah that makes a lot of sense so how do you so you guys

16:30decided to raise money yeah we only we so we only took two two checks one is from Founders Inc which is the startup incubator that I worked out of um previously build space was one of their portfolio companies as well the only reason I took money was from them was because I realized they helped me a lot at the last company and I wanted them in my corner and then we only took it from one other Angel who was Jer laav uh who gave us like 100k as well and he sold his company to beat every and he was the director of Music at meta strategic

16:56strategic right we we haven't needed to raise in terms of for monetary reasons it's more been for connections and uh yeah that's it's the only reason to Ray really and and they came to our us so it's like of course yeah that's the best time to raise interesting okay so how long do you think that will take you guys to build where it's like the MVP for the B2B side uh so we already launched a a beta with one customer yesterday and we started working on it two weeks ago so the cool thing about us is we work really really quickly like we have a team of four people and since as

17:25a startup we kind of have like very high expectations on how how much of how much code we're going to write so like every day we're shipping some new product feature like for example in the next uh 3 weeks we're going to ship a whole new redesign of the entire site which is going to be pretty exciting so we're always working on something new this B2B feature should have three Pilots by the end of December and then by early January late February like in that in that time frame we're going to have it fully launched wow and is it creating like a really high quality audio because

17:52like sometimes you listen to these AI artists and they might sound a little muffled or like a little watery so we we like prior ourselves in creating like the highest quality vocals like that's why music producers that's why we've partnered and worked with big music producers and we've made the product to look very professional like it's not focused on like the like clown use case essentially or the the and also the cool thing about that is because we're working with producers is much higher retention like these guys stay around way longer than the people that are yeah that that are just making AI covers and AI spongeb but in that sense we spent a

18:24lot of time working on our models and kind of just making them the highest quality and we expect to launch like another model the next 3 months which takes the the quality from like let's say 95% now to 97% which will be the highest by far this is a really cool idea do you have competitors that are like moving like are big I feel like big studios might like try to create something like this so we we have competitors that have essentially gone down the angle of like let's just offer Drake's voice and Kanye's voice because of that they'll grow very quickly because now they're they they got the viral element to them

18:54but the problem with that is like okay you're going to have like all these labels and big guys that have insane amounts of money on your on your neck and then the other thing is the the business model is not sticky enough because when you're marketing toward this type of user they're using it for entertainment purposes the the user that's using something for an entertainment purpose and the user that's using something for a business or or work purpose is two different type two different people there's one that has very high retention and one if the site goes down tomorrow they they'll kind of disappear yeah and that's much

19:22quicker to make so you might not really have as OTE as far as the business essentially it's just a viral product which which you can make you doing right now but you're transitioning into the go B2B gaming side that's like your big exactly so I mean we're going to have this product at the same time like this music focus product which is music by but at the same time launches brge company which is going to be focused on this gaming aspect how would you go about getting customers like as far like so so what we've done is the two people that invested in us um founder

19:50zinc the portfolio that the the investors that invested in US was this guy named Huber he founded curse gaming uh which which was pretty much a Discord compan competitor at the time a long time ago so he's he's friends with pretty much everyone in the in the gaming space and then jerus love also from beat saer they're also really good friends with everyone at meta gaming uh they they invested in like five or 10 other gaming Studios so because of those two connects we can kind of get like our first 1020 customers and once we get a lot of learnings from those 1020 we can kind of scale from there scale out your

20:19sales process interesting how would your pricing model scale cuz for B2B customers so for B2B customer it's a complet completely different game with these contract sizes like tens to hundreds of thousands dollars a year uh there's much different than consumer so we'll have the consumer angle running but for gaming we're going to be talking like we're talking to a studio right now that's that spent $50,000 on a sound agency to just do all the sound for the game imagine we could get them for $30,000 to have their own in-house tool that can kind of gener whatever audio they needed and at the same time uh the

20:50sound agency doesn't do voice characters and stuff like that they have to hire another voice actor and all that imagine you can also do that in the software as well is this month or total to like uh per year per year wow so $30,000 per year to get basically unlimited sounds that you own the rights to and then also voice actors so that's that's how these agencies kind of work these game studios they pay like per year contracts or per game contracts um but yeah we we want to license out per year so the big Hollywood strikes that were going on have you have you been like involved in

21:17that in any way uh I mean I would say like one product that like we could one way we have been involved is like the voice since the voice of voice tool kind of works for anything you can use it for we have a bunch of users that even use Music fi just to like convert their voice into like a Morgan Freeman style voice for a movie trailer or XYZ so in that sense you can kind of replace voice actors right now with our tool that sense there may have been a little repercussion but that was pretty much it interesting okay but you're in like this

21:43the AI trend is super like popular right now especially like right when you launch probably like the peak of it yeah and so I'm curious what has changed since like when you launched this to now eight months later as far as the AI landscape for voice cloning I think there's a lot more tool specifically in voice voice cloning I think like 11 Labs has definitely like cemented itself as like an industry leader but that's in text to audio that's their main focus one specific TTS right Tex to speech um there's not been a specific industry leader I think maybe us for audio to audio which is pretty cool um I'm

22:15excited like overall in the AI landscape because I'm from SF like I spent a lot of time with all the builders I've seen a lot more people kind of come in uh like eight or nine months ago we would see a lot more web 3 projects that were still like kind of hanging on but as soon as like ch and all these other apis came out people are just switching their entire companies I know companies that rate three four5 million on o 3 idea and I just pivoted straight to AI after the chat launch so it's pretty cool to see how how quickly people are what do you

22:40think about people that made that pivot I think it makes sense it's like it's it's Trends right like people want to catch on to the next hype wave but I think if you can build a really good product right now uh you can make a lot of money in the future uh like there's there's a lot of spaces in in the AI vertical that need to have cemented industry leaders which there aren't and if you can cement then you get you think it's too late to get started on one of those I don't think it's too late to get started if you have distribution like that's the only thing if you have

23:06distribution you can kind of be any company what what opportunities would you pursue if music Vibe wasn't the main priority um so right now I'm really good friend we're actually doing this so Daniel you know Daniel and Musa is another friend of mine who's a big like Instagram Tik Tok Creator so in musa's whole content is like hey here's how to become a Tik Tok Clipper and uh use your content take a take a streamers clip put a GTA clip on the bottom and put it on Tik Tok and you be created by the Creator fund we're basically building a product that kind of takes long form

23:35like twitch streams and kind of creates Clips like that very quickly with generative products so that's essentially what we're working on and the cool thing about that is since Musa has so much distribution in that type of audience if we build a tool for that you kind of have guarant customers huh what's your Tech stack um for Dev work we spend like a lot of time with type typescript JavaScript we use next as our framework and we also use super base as our database and then we don't for music F we don't use any apis like everything we run locally on modal which is a serverless GPU company yeah

24:06and why do you make that point uh so apis you can kind of just take another company's product and kind of just call into your own app there is actually no other company that hosts the model that we use which is the so we had to host it ourselves and because of that we have a lot of companies that reach out and they're like hey I need I actually need this use case of audio to audio in my app how can we get it and we have to offer them API so you have an API exactly we have a closed s API that we give to like specific customers like we

24:31have a spend limit of like uh you have to spend at least 10K and you'll get access to it but yeah that's cool so people want to basically insert the features you have in your app into their own software some way one way or the other they're allowed to exactly 10K minimum B2B basically exactly interesting that's really cool you got a good business brother nice yeah but it's so interesting that you probably didn't even see this coming you said that you were starting a new app every month yeah so I didn't I didn't expect this at all like uh the month before I built like an

24:56AI image app so so like if you remember like lenza came out which I basically built lenza two weeks before which was really cool so I I spent a lot of time just working on these new models that were coming out and and just kind of launching apps as fast as I could that was my entire goal so you did you just pick up on like these AI apis a little bit before the mainstream because you're in San Francisco and because everyone was talking about ITA I think I think it was a little bit of that and also like I would spend a lot of time on archive

25:21which is like actual research papers go on really yeah AR I was going to ask you that yeah so archive is literally like so a lot of people will go like social media that's fine too you'll kind of see where Trends are going but the earliest you can spot a trend is like when a research paper about it comes up right like a research paper about vocal style transfer came out and I was like okay this is really cool there's probably something here and then this AI Kanye video came out when you have those two connects you can kind of be like okay industry advancements are

25:49happening distribution is happening I build a product there's probably something going to happen right so an arch what's it called archive.org or uh it's arxiv it is yeah wow okay and then so basically can you see if it's getting like I'm assuming a research paper comes out and then you get is there like an uploading system like Reddit or how do you like see uh you can kind of just read it the if you want to go for like an uploading system the best place to go is Hacker News Hacker News is like the best that's kind of like archive they take actually a lot of stuff on Hacker News kind of

26:18points from archives archive stuff archiv is like literally pure research papers but how do you just are you were were you just sorting by like new and like every every day there's a couple new papers and sometimes there's some really cool ones like you'll see like hundreds hearts and okay cool this so there's Hearts like you can like sort by how popular they are basically okay cool that's a really you actually just really talk and then hugging face is the other big one too right so really I yeah hugging face you can see like the like hugging face is purely made to figure out which models are new and upcoming

26:48like so ranked is archive specifically for AI archive is no it's for everything it's like pure research papers hugging face is purely for models which which is AI and then hacker uses purely for like Tech stuff that's really cool happening M okay how do you spell archive uh a r iix i v perfect okay I'm going to look at that tonight I'm pretty excited now got my I'm all nerding out but okay so you basically have the plan you were making a new app every month yeah this one how many did you make before you made this one I think I made five or four four or five yeah okay so you made

27:21five this one hit it 100,000 users like almost over in the same first second week that's crazy so you were you were you in built base at this time or you left uh no so I had left so this was five months after I left buil so you left buil space one app a month for five months then you launch this one and it hit that's exactly what tweet Hunter if you know what tweet Hunter is yeah I know tweet Hunter IBO yeah so that's exactly what they did they were making a new app every single month and this one hit tweet Hunter Head they sold it for 10 million so

27:47interesting to see those comparisons but yeah so build space learned probably the foundation yeah kind of how to run a company how to like operate and also like how to lead a team that was the main things did you learn how to code then too I I spent a lot of time actually while working at the company learning how to code as well so I I learned how to code in the span of like a year and a half entirely from no code to like I know how to actually build products and you feel like you're like a pretty strong developer at this point yeah yeah and that was purely just cuz

28:12like I I was like I know that if I want to create like really big products I need to understand the products themselves like there's a lot of talk about like hey you can use new code platforms but the amount of custom you need for building a great product you need to understand what you're doing so I spent a lot of time interesting we're proponents of no Cod but I totally understand what you're saying yeah so are all the people on your team you said you have four people are they all developers exactly Engineers so four Engineers I would we have four or 5,000 Affiliates that do content for us oh my

28:40God yeah so that's the other thing so like I wouldn't say they're on our team but we have like this one 16-year- old kid that kind of manages on it's pretty is it just like in a Discord server yeah it's like this like this kid this I paid him $500 a month but from the beginning like this the the month that I launched it till now he's kind of just been managing all of our affiliates for us and he's done an awesome job that's awesome was it on Discord though yeah it was on Discord and what do you guys use for your affiliate prr we used a

29:03software called tol tt. why do you use that one so it's great for essentially uh normally for referral links you kind of have to like set up emails and like you have to send an email verification then you have to approve it like we didn't want to deal with that we wanted people to kind kind of come on uh log in and then have their own link desde their own link whatever it is and and ready to go so there's a soft software called to which makes it really easy for people to make their own affiliate links I think it's one of the coolest companies in the

29:30space right now not a lot of talk about the company it's one of my friends companies but they're doing great what's like their unique proposition the unique proposition is essentially anyone can come on and make an affiliate Link in less than five minutes and track and they also have a dashboard to track like leads referrals like leads as in how many people converted which plans they converted on and then also referrals how many people clicked on your link and didn't buy anything and how people signed up so really good dashboard is it mainly for software or is it anything you can kind of use it for anything that

29:57has a stripe checkout so it literally uh hooks onto the stripe endpoint at the end but yeah that's really cool okay I'm probably going to use that yeah tols is really good like I really tt. it's I would say one of the best companies I've seen in the last year okay great so left build space new company every five months this one hit 100,000 users your distribution was basically Affiliates like went viral kind of you kind of caught a trend early I would say yeah then you had Affiliates to put fuel on the fire then that's kind of established you gave you the capital to think bigger

30:29you have did you have four people from the start or did you bring no it was just me for like the me and like one person for like the longest time for like maybe four or five months and then we slowly brought in more people because we're like dude we're just like saving like so much money every month like we should start spending it you know yeah no cost and then you brought in so you got two more developers are they partners are you the uh yeah so I'm I'm the founder and then we brought in two like part-time like Engineers essentially gotcha great so you have

30:53four people had some money then you brought in two strategic investors cuz you had had the bigger B2B play in mind going after gaming so you got the beat saber guy and then founder Founders Inc Founders Inc startup in cool and so that was literally just to help so they could open doors give them a little piece of the company and then that is your vision for the next year is just build out this gaming side and yeah I I in unison I think it's it's two things so one is like I don't think you can build a billion doll company in music right now and the main reason is because not

31:21enough people know how to M make music and because of that like that essentially comes down to one main thing which is like these music softwares are really hard to use for example if you look at like Adobe Premiere a lot of people can't use Adobe premere that's for images right but when mid Journey came out and allowed anyone to make images like the market for image and art grew like 5x if you do the same thing for music which allows anyone to convert their voice into instruments or voice into other vocals and make music with just their voice or or simpler tools you can kind of grow the market in one way

31:49so that's one part of the equation we want to grow the market of music for anyone to make music at the same time we want to start building out the gaming aspect interesting that really does make sense that you the Tam will grow because of your technology exactly more people feel like they can make Quality Music yeah and then the gaming side makes tons of sense music is like 50% of the experience so you'll save them a lot of money and time it's a no-brainer so are you busy like yeah I mean I spend a lot of time like working to be honest like I think I

32:19I just got an uring just because I like I sleep so little nowadays like really yeah yeah I spend so much time kind of just working on random stuff I'll wake up in the of nights literally sometimes with just random ideas that I'll have to write down and this may not just be specific to music VI this may be like oh I have this friend that has this crazy like for example I have a friend that owns the domain stocks. and like last night I woke up with the idea of like dude if he owns the domain stocks. a I have a friend who's like one of the

32:44biggest stocks influencers imagine we just buil a software product that kind of helped people analyze stocks and just interesting and it's just like random ideas kind of just come to me so what ideas because you I'm sure focus is like probably hard biggest thing you have focus on because you have like now you have success a lot of people are gonna come to you with like a lot of cool opportunities that a year ago especially being like the one of the only devs in the kind of Creator space a lot of people come to you with like ideas like hey I have like five million followers I

33:11want to build this can you help me like a bunch of inbalance so you have to kind of stay really focused how do you justify like you said you started a side project with some other people yeah so what so what I'll do if I do anything like that is I'll basically bu bringing my connects so since I'm a Dev I know a bunch of Developers so I'll be like okay if you want to work with me on something my my friend XYZ is gonna work on it I'll kind of just hop in once or twice every month but that's it and and it's been working great so that's what I'm

33:37doing with Musa that's what I'm doing with a couple other people as well so you're kind of like the project manager but exact finding the talent and over and at this point it's like I don't really do anything other than when they ask me for something but then I'll take like a percentage in the the project what are some cool model or cool research papers that you've seen that would be interesting products but you just obviously don't have the time or like a random idea that' be great but I can't I mean mistol came out with a new model like last week right mistol is like the new text to text model that's

34:04almost as good as GT4 but it's all open source like you can download the weights on your own computer and run it locally so with that you can kind of do anything and it has Apache License which means you can use it for monetary gain um so you can build a bunch of products for example one of the ideas I think I would build right now if I had the time was a mobile app that runs the seven like mistrals model so essentially now when you have no internet no Wi-fi no service you can still have an llm running on your phone locally uh to do whatever

34:32essentially so it's basically an offline large language model you can like download to your computer exactly and if you if you build an app around it right let's let's say $5 app that people can download and that means they have internet access forever right because essentially it works without it runs locally on your phone didn't stability try to do this or it's just not as good it's not as good m mistol is by far the best model that came out and it came out literally last week so this is like brand new m s r l uh m i s yeah mistol mistol huh an Apache License just means

35:02that they're allowed to monetize it exactly like for example llama it's like Facebook's model but llama doesn't have an Apache license so you can monetize it legally so if you built an app with llama's model running on a phone you can't mze it are there any like good text to video models yet like that You' seen uh there's a couple I mean there's a couple of really good lip sync models which are like take an English video and turn it into Spanish um so those models are getting really good which are called wave to lip models text uh texted video models are not great yet but there's a

35:30company called um uh pixel which is making really good models as well and there's another company that doesn't do text to video but does video animation which is called Wonder Dynamics I don't know if you've seen it but it's essentially you can take a camera and kind of record someone and mask them with a character and it's a really high definition really good um but over the next year like those are the models that so like text to image kind of improved the most this year but over the next year text to audio and text to video are going to improve a lot as well yeah text

35:57video seems the most interesting to me just because it's going to like change the ad creative space yeah I mean mid Journey's already done it I would say in a year from now like there's going to be a lot more fake videos and fake media out there as well just because of how much text video you have any idea how to verify deep fakes or to stop deep fakes so there's there's a couple of companies that are now spawning and that's one of my friends built this company called Deep fig which kind of just detects like if or it's called Deep truth they detect if there's deep fake audio um but it's

36:22very hard to detect still to this day how could you do you yeah it's very it's very hard to the human here you can't tell but like if like as models get better there's a lot of talks in governments that hey we're going to have to start labeling outputs of these models with some sort of like uh graph that we can understand and Mark out as this is not real this is real because there's a lot of political issues that are going to be caused there's just no way that there's just no way that you're going to be able to hold people accountable to that like anyone's going

36:46to be able to copy the model and make one without that and like the the cat is out the bag right like all these open source models are out and they're never going to be deleted again yeah my theory is blockchain and nfts can solve this but we'll save that lecture for time ipfs man yeah but zero knowledge proofs I mean that's what Sam alman's doing with worldcoin so I thought if he left got kicked out open AI he'd go over there but I don't think that was ever going to happen World coin is up this week like two 20 30% B coin itself yeah yeah or World ID I guess is like what

37:13I'm referencing but worldcoin yeah it's a great project to look into but not Financial advice I'm one of the people that pivoted from web 3 to AI yeah but I'm still I think web 3 technology is the most important like technology the the biggest problem that happened over the LA so before build space I used to work at a web 3 company we were building like crypto ETFs essentially and really yeah so like when I was 15 16 we raised $2 million and the the idea was essentially like in in the S&P 500 you can invest in one stock and invest in C on the entire Market we want to do the

37:45same thing in crypto invest in it's pretty big right now yeah that's the narrative right and um so it was like the idea was like basically building an ETF the problem that we came to realize was like right now or like back then even until this day now the the problem of adoption is still the main thing it's very hard from a non- crypto person to become a crypto person and when and and the people that do become crypto persons they're like really deep into it you know there's no like middle ground it's either zero or one it's like you have that light bulb and then you're like a

38:11maxi yeah either either you're like I don't invest in crypto or you're like oh yeah salana ethereum like xxyz I have these nfts right yes I I guess people are too harsh on pivoting from web 3 to AI the reality of the fact matter is you need to make money and you need to survive and right now it's too early to the technology it's like trying to create an iPhone when you don't have Wi-Fi like it just won't work age and so once the technology scales then I think nfts are the most important technology in the next decade for a variety of reasons I won't go into but I think it's

38:38too risky I think my billion dollar business I already have it in my mind it's blockchain based but it's just too risky to bet do I build on ethereum do I build on salana do I build on madic do I build on avac Cosmos you don't know which one's going to hit Mass adoption and so there's too much Market risk right now to bet on that yeah I think I think once once we see more of the mass audience kind of adopting the the tool then it's a UI problem I think uiux over anything too many people get drained so just too early for the tech too slow

39:06blocked no one's paying $10 for gas fee for coffee so yeah I mean actually I I was a big crypto Maxi but then when AI happened I switched over so I used to actually go to like salana breakpoint in Lisbon I went it I went for two years in a row so like I've met with Anatolia and Raj and I was a solid de before that so I was super deep into web 3 like that the reason I pivoted was cuz I realized like okay cool like there's this problem of adoption like like I should probably find a way to make money really quickly and all these new models are coming out

39:32so it made it really easy to kind of just move over CU everyone is adopting it way quicker and it seemed to have like much more utility it's ready today there is actual value that applies to every business immediately now yeah but four years ago the longterm effect I think of crypto and web 3 is going to be way bigger do not so I think that this since you I didn't know that you were into web 3 that's awesome so I'll actually nerd out with you on this I think that uh the way Elon has forced everyone to get verified on Twitter how he changed it to just using your bank

40:02account and the name on your bank account to prove that you're human I think that nfts with World ID will do that across the entire internet so you have to like connect your wallet that proves that this is you a real human with an nft have you gotten scanned by the way not yet no yes I've gotten scanned wait for real what's the process so my friend Alex mazm he has he literally so he used to work at his his roommate works at um world coin or world and he has the actual orb in his living room like it's like a silver orb really do you trust it uh dude it's like the

40:34creepiest looking thing and it has like a little circle in the center it's like a black circle and he was like yeah bro just put your eye you'll make $50 easily and you just put your eye and it scans it it scans your iris and then it just sends you 50 $50 to your credit uh scary they pay you to do it yeah it's it's pretty wild and what go ahead I was saying like the ID app is pretty sick as well is it in your web 3 wallet or do they have their own wallet they have their own wallet and can you put other Assets in there or is it just like stand

41:00alone just that right now do they will they allow you to transfer it to like a metamask you can transfer it you can transfer Ito that's so what happens if you lose it do you know I I mean it's just like kind of like just for setting it up you're not like you need it you don't need the currency no but the actual ID itself oh you can't lose it it's like attached to your it's like an nft essentially if you lose it but you can't transfer wallet to wallet you can't transfer from like the the nft you can't transfer but the currency you can yes okay so the nft stays in their wall

41:28wallet there's already been demos of it they're like they just launched the ability for you to connect your to it authenticate it really you're like yeah they launched a bunch of connections like last week see I think that's like one of the most important Technologies right now because that's how you stop deep fakes if you make everyone connect their real identity to their accounts on social media you're going to think twice about what you post and if you post a deep fake of a presidential candidate doing something you're gonna get like you'll get defamation immediately and so that'll really and if it's not from a verified account you will no longer

41:58trust it like that has to be the only way to stop deep fakes so I think AI is going to like expedite the adoption of web 3 that's my assumption I think so too I think I think it's G to have it's going to be the reason that people have to move on have to do it right that makes a lot of sense is World ID last question we'll talk about this because I know this is a very Niche group of interests here but I do think there's like a it's these are just future Technologies and that is a demographic of people and so it's not pivoting it's

42:21just what's the interest right now based on the advancement of the technology but do you know if World ID is is it its own blockchain I'm not sure World coin okay I'll need to look into that then okay well that was a fun conversation probably the most fun I've had you should get scanned though I should I want to I think the only reason I trust it everyone I always see like conspiracy theorists talk about Mark of the Beast and that this is like whatever they're going to steal your data but the zero knowledge proof side it makes sense you know what those are expl explaining it is like the hardest

42:49thing I would say like the example that people have given to me is like imagine you could prove something without actually showing it's so like the exle people use is like imagine you have a number behind like a whole string of numbers behind like a wall and like you have a small circle in the wall and you can kind of just see the one number but there's proof that the rest of the numbers are there and with that there's proof that there's something behind the wall and it's Iden identifiable it's just it's hard to explain but that's kind basically put you put you scan your eyeball in here and that eyeball is

43:19associated to your ID but then it goes through like a mixer and then no human can see it's only like one way exactly it's like it basically the system knows what is what so it can verify who is who but no human can look at the backend you can't go backwards exactly you can't you can't see whatever zero knowledge proofs that's the reason you want to trust IDs and if you want digital IDs you want nfts and you want zero knowledge proofs not from the government okay just making that very clear so sorry no worries back back to your back to AI software because that's what most people are interested

43:54in right now so what if like I just like giving people ideas on like what you would what are some of the most impressive models so you said mistol right mistol is really good I think there's a bunch of like even stable diffusions model like um 2. 2.1 came out which is really good for text text image I think all these models if you if you have distribution for a specific angle let's say for example I'm an influencer and I have like a huge like platform that has a bunch of followers about dogs right like I post Instagram pictures about dogs I have like a million

44:22followers that's my entire account I would maybe build like a a text to image generator of the custom pictures of your dog right like using stable diffusion's base model hey we'll fine-tune your dog's images on this and you can now generate your im images of your dog in Paris in in France wherever you want you know like building these custom use cases for these open source models with your specific distribution angle attached to it interesting I think just combining all these in unique ways is a big opportunity now do you think that like these are like very shortlived do you think open AI is going to like

44:55Amazon on Basics all of these people and take it away uh I think like a lot of companies would die tomorrow if like one API got turned off like for example um if OPI died tomorrow like I would say like 30 40% of the startups in the world right now like AI Focus would just die because they all rely super heavily on one API the other thing I I do do meant to mean to say is like I don't think that they'll just disappear tomorrow I don't think openi is going to be able to build like these really specific use cases for example I have a friend that

45:24that builts like a an sell plugin that's used for drop shippers to like kind of generate scripts and and thumbnails and and a bunch of things for essentially their drop shipping stores using like open a and stealable diffusion and a bunch of products and like it's a very specific use case it's for drop shippers to generate like uh categories and generate the the titles and prices and images of new products you can really like open is not going to go build that as a new feature right like these very specific use cases I think there's still going to be companies that are going to be ried on their API and build really

45:55really good flows and then if you do that you can make a lot of money still too interesting okay how do you see the AI landscape changing like Google search like it seems like I think open AI you're in San Francisco so I assume you actually like see more of the stuff than me so they said they wanted to make a phone correct like that's open AI yeah yeah I I think I think so everyone said that and so do we see it just becoming like like how does SAS exist if opening eye makes it like a phone with no app app it's just like you're talking to it

46:27through the chat gbt interface and then whenever you ask for a function it just automatically routes it towards a SAS so do you see like basically what I'm trying to ask you do you see the SAS Market growing meaning there's more room for more micro apps or do you see the the SAS Market shrinking meaning less apps are doing more work I think there's definitely going to be a lot more micro apps being built just because like the entry is now way easier like the entry into that market is way easier you can use a lot of these L llms actually to help write code with you as well and

47:00also you have bubble and and all these other like no code tools as well that can help build tools um I think micro tools are going to grow like 10x over the next year just because I know a lot of my friends that didn't know how to code a year ago and still don't know how to code as much today but they're now able to build like really simple apps just with like the help of AI or like tutorials on the internet using bubble um the that being said I'm not sure about like the bigger bigger companies I think there's going to be a couple of

47:25really big industry leaders that kind of solve a lot of problems like for example open ey is one of them right solving a lot of the text to speech or text to text problem they even solve text to speech they they launched a new text to speech Model A couple weeks ago which is like battling 11 Labs right so there's going to be companies that kind of just take over whole Industries just because they have one thing that others don't have and that is compute if you have gpus you can kind of do anything open a has the most gpus by far of any every any private company right now and

47:51because of that they can train these really crazy models and also have all these apis available that don't crash because they have so much Compu computer is the biggest problem if you have a computer you can build a big big software company and a lot of people won't be able to get that much Compu right now and because of that they'll still there's there's GNA be way way more micro apps than bigger apps okay so if you had to break down your philosophy to someone who's just trying to break into their first sass because you're in San Francisco your success at your age is absurd by the way so really cool to

48:21meet you bro but the fact that you started so young you're in the right environment you went through build space you're seeing like all these best practices from the get-go how would you suggest the everyday person maybe a 20-year-old that's just getting started they don't know how to code but they want to start a software now it's more accessible yeah I think the playbook for that is really really simple like if you don't know how to code I wouldn't go say go learn how to code I would I would first go say find someone that has distribution because if you have distribution you can sell anything like

48:49if you have a million followers you can probably sell a really sick like dirt dirt brand you can literally sell anything that's the craziest part so go first find distribution this is what I've been doing for all my new new projects then I'll find like once I have distribution locked in let's say I a I have a friend that has like 5 million followers that does music content then I'll go find a developer finding a developer is actually easier when you have something that they want which is distribution all developers want something that they can just get to 100,000 users very quickly and when when you have distribution you have what they

49:19want uh find a Dev that's going to be willing to work on your project with you in in SF because of who I am I have like a bunch of connects in that space But if you can go into like a bunch of these Discord communities you can find devs that are working on random projects that are kind of useless and you can just be like hey look if you want to become a millionaire I'm going to help you do it I have distribution locked in I know how to build good products if you have a good track record as well that's that helps and then if you if you have the

49:43dev on your side as well you can get your Dev and your distribution to work together and that's how you can build something kind of skills right now what Discord groups are you referencing there um there's a bunch there's one called founder Founders zc has its own Discord Group which is super Dev Focus but even if you go on like Discord and look look up just software engineer or AI discords aiub is another one aiub has like hundreds like thousands of devs just in there just like writing code every day and like any of them can build apps like these in like a couple days but the

50:09thing is they don't have any any reason to because they don't have any distribution a lot of devs actually don't spend a lot of time like Distributing their products they spend a lot of time building and because of that they don't have that innate skill and there's it's it's hard to find Distributors but it's also hard to find devs but when you get them both together that's when they become superpower it's very interesting it seems like developers have this ability to build anything but they lack the ability to understand what people want and then people that have large audiences understand what people want but can't make a product for their life so you

50:37just have to become a dealmaker that's essentially it oh you're saying don't even you're saying be the middle like find a distributor find a devel i me that's that's what I do for my new projects and that's what I would recommend Lage yeah if I were to start tomorrow with no experience and anything that's what I would do I'd find someone that has distribution and then I would find someone that can build products and kind of just work on something together and leave the project I actually really agree with that it's like why even maybe spend a little bit of time learning how to coach you can communicate to them

51:02yeah just know understand yeah maybe spend a little time marketing so you know how to talk to an influencer but if you can just really see a hole in the market then go out make a list of 10 influencers that might be a good fit for this send out some messages hey I have an idea and and the cool part is devel the cool part is like all these influencers are trying to build software products right now why do you think that is well that's cuz it's like the most sustainable business model right now like does like YouTube and all these other content you're you're on another

51:28platform right you you're a slave to a platform but with with SAS or your own platform you can kind of go to sleep tomorrow and know that you have an expected amount of money coming in next month uh and because of that there's a bunch of influencers are trying to get into software and because of that if you can go to an influencer and be like hey I know you have this type of brand I can help you build a software product in the next two months if you want to work with me and most likely than not if you hit up 10 people maybe one will say yes if

51:53you hit up 100 people maybe two will say yes right whatever the the number is find the people that say yes use that as leverage to find your actual Dev you don't have to have a Dev first use the leverage find a Dev and now you can actually build something devs are quite easy to find I know that if you're not Technical and be a little intimidating you don't really know how to like find them or ask them questions but yeah everyone thinks it's very hard but it's they they all are in certain pockets of space you just have to find the pocket where does a Dev live where does a Dev

52:18learn uh I would spend they would spend they spend a lot of time on these research papers if you go on these research papers you can see the comments if you go on Hacker News you can see comments like all these places are people that are writing code and they're giving their thoughts GitHub is another one there's a like hugging face is the biggest one I would say if you spend a little bit of time on hugging face you would find hundreds of devs and then the place that I find most of my devs is on Discord I'll just be hanging out in discords and I'll be I'll drop a the the

52:43test that I usually use is like I'll drop a bug in the chat and if someone solves it and I'm like oh okay cool good to know right and then I'll have them in my like uh Rolodex if I need them for something that's actually really smart you know devs love to like Flex their ego I'm going to solve this problem exactly exactly that's so funny that's really smart good advice okay so would you say software is like like a first business like a good first business or is it harder how would you describe it I would say it's like if you kind of understand like at a certain

53:14point it's a really good business and then once you get to like 50 to 100K 200k it gets harder to scale because it requires so much more persistence and building a really good product the other thing is like solving retention problem is really hard to scale like if you make one change with the button if you make one change with like the login page on your UI maybe your attention will drop by 5% or increase by 5% So what at scale it gets really hard but that's where decision- making comes in I would say anyone can start a software company today if they just know how to make

53:43deals and that's bring distribution bring bring devs together and I would recommend it because there's a lot of money to be grabbed in the space just because all these open source models are being literally released like these months these weeks and if you can take them quickly and put them into specific use use case quickly you can make a lot of money so how much is it about launching first to capture market share and then monopolizing that space then maybe taking your more more time trying to really refine a product and launching six months later with a more polished product um I would say like iterate in

54:13the wild that's like the biggest thing iterate in the wild I like that yeah I mean even at build space we have this believe believe believing we have on our sweaters it says gtf of all get the off Local Host Local Host is essentially like when you code on your computer it's like a local computer so you're coding locally so never never code purely on your local H like when you have your first iteration of app launch it into the wild and then keep iterating on top of it that's what we did with music f i I would have never launched music F if I wanted to build like a software app and

54:41have like login authentication Securities and all that and launch it we launched it into the wild saw what the problems were and kind of just keep iterating really quick so you might have touched on this but I just want to make it really clear you didn't have distribution when you started music 5 at all like you were just a solo Dev basically so how did you go about getting those first was it 40 beta users so the way I found those people was there was the Discord called aiub and there was a bunch of people that were trying to essentially use these models to make music and I realized they

55:11weren't devs because of the way they were talking code and I was like okay cool here's a site that you can kind of use this Tech and they were giving me a bunch of feedback like hey this is great like it would be cool if you had this XYZ XYZ XYZ and in spent of a week I was like okay cool this is ready I launched it out and even the first version it didn't have any authentication like there was no login there was no rate limiting so that means if people just kept pressing remix it would just keep calling our API and that's why there was

55:36so much more cost in the first couple weeks but because of those mistakes you can kind of move quickly because you learn your mistakes in the wild and you quickly that's really interesting so you literally just went found a Discord of people already doing this musicians trying to make AI music and the way I found the the Discord was like I was on one of the AI Kanye videos and on the bottom was like um with this Discord or in this Discord some sort of Creator distributor content creator had a community around this topic already and you just want to say hey would you guys like to try this out it might help you

56:03and then and then I realized like okay cool if I'm going to build this I'm going to I want to have at least some guaranteed distribution so I just told the founder of that that Discord I was like Hey I'm going to build this out um it's going to be really useful to a lot of people in this Discord it would be really cool if you made an announcement about it and he was like sure so the week down yeah and and I think that was the biggest like push because as soon as he announced it like we had like 100,000 people or 10 10 20,000 inter quickly go

56:27to the site that's so smart yeah I think a lot of people look past this but how we think about it if we have a software idea immediately we're going to create a Discord Community around the topic that people that want that Discord server want we're going to create a newsletter create media assets Instagram accounts Tik Tok theme pages around those topics and then you kind of just present your product naturally as you're building it but you kind of need more have you ever tried to build like distribution yourself or did you just like build my own so I did when I was younger and I

56:54realized like it comes way more down to consistency and I just didn't have the time it's a fulltime job yeah it's it's like and it's it's also repetitive it's like doing the same thing over and over again like I wanted to solve different problems over and that's why I think Dev work came more fun to me because it was like every time it was a new model or there's new something new to kind of do so because of that I really liked that like I think there's two types of people some people that like Distributing more and some people that like working on products more I'm interested now that

57:18you're at that 100K Mark and you're going B2B how would you that does that kind of Contra not contradict but it's not this it's not in line with distribution as much is it or it's just a more no so this is more like hand toand combat more dense um but the the reason we want to do that is is like I mean we're going to lead distribution still like we're going to push harder on distribution I'm just saying my main F like we have a team of four people so like one or two people are going to purely focus on the product and launch the redesign launch more

57:45distribution but I'm focusing on this more specific product which requires me to go into meetings with other CEOs and Founders and kind of be like hey look this is the product this is what we want to do and I want to get learnings essentially I want to spend the next two months kind of see if there's a market for this because I think the market for that type of product is actually much much bigger and that that's where your exit exit value can come from B2B deals because they're not going to once you get that client they're never leaving basically yeah essentially like these these Enterprise yeah Enterprise clients

58:12are much more attentive and if you can get one or two really quickly that's really good feedback but it's a much more upfront sales process it's hand to hand combat that's literally so how would how do people in build space or these other people starting their own s with no distribution what are like the best ways to Market a SAS um there's like so outside of cont like yeah give me like three or four I'm like actually really curious outside of content yeah if if you don't have a content creator influencer someone to push it organically are people running Facebook ads are they using I have never read any

58:41ads for music f for any so I can't talk about ads but what I can talk about is like if you can get other people to have a reason to share your product basically yeah so I think that I feel like it's was a huge unlock like people don't really realize it until they kind of do it even uh my friend Musa who does like the he has his info product as soon as he launched uh Affiliates on his info product in like 3x so it's like Affiliates actually huge power and people people have like a frowned like reason to use it they don't really use

59:09it that often and I wonder why but I think Affiliates is a huge one if you can understand Affiliates and provide someone a reason to share your product right that's a big thing if you can understand like what a good product looks like and feels like for most people like for example uh early on we've seen like a bunch of these AI products that kind of it's like super Jank and they have a bunch of bugs and if you actually just make a simple really good product that actually does what it intends to do you'll have good distribution just by people talking about it right so that's like the other

59:36thing so it's like Affiliates and having a good product and I would say the other thing is like even on LinkedIn and like having really cool demo videos like for example one of the demo videos that I released on Twitter it got like 4,000 likes was I took my voice and I would I would just like hum a rhythm into the microphone and I would convert it into an instrument so it would be like voice instrument like voice to trumpet or voice to and it was super viral on Twitter it got like 4,000 likes and stuff like that so very viral demos are really cool too so you have if you have

1:00:03a novel idea and you can express it in a demo and to show people a cool use case yeah the first demo I did to launch in the aiub was I took my voice and I was like hey you might might be wondering why I sound like Drake right now and then the second a second later I was like now I sound like Ariana Grande and I replaced my voice to sound like Dr aric Andre and it hooked people in right like they're like okay cool what is this Pro product yeah I saw verse do this an Instagram page we did a really good job of like talking like that and it went I

1:00:30think got like 30 million views overnight so it's like just create really good demos that's one thing if you can if you don't have content creators and all that just if you can create demos if you can get Affiliates and if you can just build a good product you can kind of get out there really quickly you need like this critical mask and then if you have the right pieces in place it can people people like for example now like a lot of our content we don't do ourselves like like we have 10 or 20 YouTube videos that we didn't make like big content creators just made

1:00:56because our product was good and and also the affiliate link kind of just pushed them over the edge to share it okay dude I've learned so much already in this hour this is awesome so thank you where can people find you just right now they can find me on my Twitter um ARB K24 and then on my Instagram RB K7 a r i a ib K7 yeah and we're actually hiring right now oh yeah where do they go so if you're a content creator just hit me up on Instagram are you hiring content creators in house yeah so we're working on something that requires like a bunch of content graders to create

1:01:26content about our product we're doing something similar to Oliver uh from oh nice hell yeah dude he that was like he shared the sauce on that one so shout out to him because that is the strategy of 2023 but so you're hiring content you're like basically just like in a Discord group and you're gonna say hey make content on this subject are they gonna be Affiliates or you actually going to pay them like well we so we so what we do is we take our biggest Affiliates and kind of just give them like monthly or weekly Sal cool and so you're also trying to create create a

1:01:55personal brand for yourself yeah why I think like in the next like five years if you have like a personal brand you can kind of sell anything that's why you see all these influencers sell like like info products like info products have like no cost almost like almost no cost and then they can just sell them for INF am infinite amount of time I think for me I kind of want to use it as a ability to kind of launch products really quickly and have like that first zero to one very quickly so if you have a if you have your personal brand and let's say I

1:02:21have like 10 20,000 followers and I'm like hey I'm launching this product you kind of already have a guaranteed 5 10K us which is really cool so that's that's my reason also just like you kind of get to other things in terms of branding and and distribution like you can talk to other people 100% and it kind of gives you like you kind of very clearly can provide value to someone that's what George was explaining it as like if I walk into a room literally just at any moment if I say hey come on the channel it's like instant value out so people kind of want to like work with you work

1:02:50with you and it's easier out the gate and there's no words needed so it is a superpower and yeah I can we can launch any product get like a testing pool right out the gate so huge advocate of personal brand obviously over here but I do say I would say that courses are probably one of the best business models but you do kind of have to earn the right to sell a course like you can't just like make up a course like you have to practice and do something but you're definitely in the position where you could I don't think I'm going to do a course but like my idea is like I want

1:03:17to create a a community of like content creators like these Distributors and these devs and kind of just help them make products and then kind of just take a percentage in these products kind of like kind of like what build space is doing in a way yeah build space doesn't take the percentage oh okay are they going to though like come the plan they don't plan on right now like as do you have any idea how that monetized I guess you probably should comment on actually no like I used to work there so before we were actually the world's biggest web 3 developer ecosystem and so we got paid

1:03:45like millions of dollars a year by like hey to deploy this node use Quick node or hey use build this no no like as in like yeah education as in like okay if you're going to build n project build it on salana and we have coures on how to build on salana and salana would pay us like half a million and St like that I want to talk to them because we did something very similar yeah so we we that was what build space was and then we did pivoted away from web 3 towards Ai and then now just towards any project and like the idea is like anyone's going

1:04:12to be able to like we're going to figure out a business model in the future I think that's the goal but the main business model right now is like we're getting paid by uh governments and subsidized to have subsidized education for like the highest performing people right so it's like a private school for like the best of the best so it's application based yeah so we're they're having inous in-house um apartment so we have like in San Francisco we have an actual campus which hosts 30 students every year every three months yeah three for three months yeah for three months that's cool so we just did the first

1:04:41batch we're about to do batch two and it's completely free for the students the government based so that's the plan and now they're going to launch one in Dubai but the online program is free so it's like a real school for entrepreneurs yeah it's like my passion right there so that's awesome to hear I want to talk to them that's really cool okay so I want to go back to the course thing because I actually Russell Brunson you know who Russell Brunson is funnels clickfunnels okay every developer if you're trying to learn marketing Russell Bruns is the only book you need to read Doom Secrets but his fourth book The

1:05:08Lynch pin he started clickfunnels it's a billion dollar company from zero to a billion in four years I believe yeah and how he did that was combining SAS in courses and so to Market his SAS platform he created a course on how to run your business yeah then within the course it's his software that's essentially what we were doing at build space but we were pushing all this traffic towards like salana and these bigger companies that makes a lot of sense it's like if you had your own crypto and you're were pushing like hey here's how to build the best nf2 project right now and it's pushing to your own

1:05:36chain that's like the same that's what we did but we didn't own ethereum but think about Drop Shipping with Shopify like Shopify I don't know if they came up with the term Drop Shipping but now there's a million people who sell Drop Shipping courses and every one of them Shopify and now their affiliate system is going crazy yeah I I think their affiliate is like $500 per or something like that it's I think it's yeah yeah I know what you're talking about yeah it's crazy because I've talking to like some really big Drop Shipping people and they're like dude like that's how we make our real money like that's that's

1:06:04that's the sauce those people but they theyve been around for so long they make so much money that they can pay like the entire LTV up front to Affiliates yeah but I just think that's a really big unlock for a lot of people trying to get into SAS is that like I think a lot of people have a stigma on selling courses but the smartest thing you can do as a SAS company is sell a course teach people a business opportunity your sass is just like conveniently placed as a way to execute that business basically so I think that's really smart marketing but okay I do want to ask you now I so

1:06:34we kind of asked you like how the landscape has changed over the last eight months with AI how do you see the next like how have you seen your approach to this stuff change as you gone through as an entrepreneur uh like in terms of building products and stuff yeah like when you just started out you really didn't know this was happening what is like how have your priorities changed I think I've come to realize like there's a lot of hard decisions you have to make yeah that's like part of it I built space there was like a like I learned a lot from Fara farza gave me

1:07:02the saying he was like Hey if you have a decision like decision a and decision B that someone gives you um most of the time as a Founder your job is to make a decision like Pick A or B but when you when you've G when you're given two choices your brain hyp like thinks like okay I have to make a choice but sometimes you just have to ask yourself why can't we do both and like stuff like that there was things I had to learn where like sometimes like your brain just convinces you like you have to make a choice but why not do both and then

1:07:27your job as a Founder is like figure out how to get the supplies to do both uh the other thing is like if you don't make decisions you don't move at all you're kind of stagnant that was another big thing we spent a lot of time kind of learning how to scale as well like when you're scale when you scale your product from like 10 users to 100 users it's like a huge thing like you have to actually figure out how to supply 100 like for example we were using we run on gpus so we had we had this problem constantly of like okay we have 10,000

1:07:52100,000 requests to to day like people did 100,000 converg today how are we going to get enough compute for that so it's like solving a bunch of these problems on the Fly while your product is out there puts just added pressure on you how do you yeah so it's like that like the fact that your product will die if you don't solve them kind of makes you solve those problems you know so it's like okay I mean tomorrow we're not going to make any money if we don't have more computers so how like you're going to figure out a way to find more computers is that I'm not a technical

1:08:16person but is that just AWS like you're just like we don't use AWS we use model model essentially like normally some people pay for gpus let's say like want 10 A1 100s do you know a100 yeah Nvidia SC yeah right like really big gpus but the problem with that is like let's say you're only using five at a time sometimes at at night you're using eight and then sometimes you're using three so you're not utilizing 99% of it so then there's a thing called serverless which kind of essentially let's say right now you're having 50 requests so it'll spin up one GPU let's say it needs three more

1:08:45gpus so we'll just spin them up and when they're they're done being used it spin some down so that's what we use so we utilize our comput okay that makes smart thank you serverless gpus serverless okay cool so let's just go into all the software that you use to run your company now all the like internal tools you're using communication organization everything awesome yeah we use Discord still internal internal stuff but I would say there's a a bunch of tools that I recommend one is post hog post hog is by far the best analytical tool out there and it's open source I would recommend using it if you if you have

1:09:19any product even if it's not even software product you can it's one line of code that you add to your your site and you can see heat Maps like video recordings of what your users are doing you can see where they are in the world what device they're they're on like a bunch of these analytics how do you spell that post post p h h o g post hog okay post hog really cool company I would recommend using them then we also use retool retool essentially what it does is you have a database that has like your users or your conversions or your subscriptions and a retool can make

1:09:49really nice graphs of what information is in your database so you can see how many new users you gained how many new subscriptions you gained and all that retool retool we just move from stripe to paddle paddle is a different POS system the reason we moved from stripe to paddle was we were seeing that uh we we couldn't accept payments in p uh in PayPal and PayPal is a huge thing worldwide payments are done a lot through PayPal and we couldn't accept them so we move from strip padle and because of that we've seen like 20 to 30% increase in subscription conversion wow yeah so paddle is another just padle

1:10:21p paddle.com paddle.com they're actually built on top of stripe but they just add more features oh okay so so tot still works for you then yeah tot still and tolt actually has they have two options they have stripe and they have paddle so have both options um so yeah I would I would use that I would use super base for your database super base is like a really good type of format for your database you can see your tables really cleanly um yeah so that's like an air table competitor super base or super base is like literally like a pure SQL database but it's like all hosted in like a nice way

1:10:51got um yeah kind kind of air table more like a actual tables but Bas actual database but you can see as in table format like air table gotcha um and then yeah I would say there's a couple of other ones um I think the those are the main few that I use every single day and then like for productivity I use like Kon Kon is like a great calendar app I use um Kon is great for calendar and I then I also use superum Y superum is like the greatest tool and it's like that's what you call a great product So like um there's a saying that

1:11:21uh someone once told me and I he I was like hey what's the difference between a great product and a good product it was like a great product is something that if I give to you and I take it back you're going to fight to have it back and a good product is like if I take from you and I have I take it back you're just you're going to complain once and that's it like superhuman is something that if you take from me I'm GNA I'm gonna fight to have it back right like I can't go back to normal emails super human what's your favorite

1:11:43part about it I just think the Simplicity like the the ability to see if other people have seen your emails uh the ability to kind of just send emails on the fly like it's not a text message like all that is just like and the whole interface I'm so used to it now and accepting Meetings onl online really quickly all that can you turn off on the other side if people can see if you opened them like on text you can turn off read receipts uh you probably can like it's in the settings but I I I know on superum you can see when other people

1:12:10open your emails and exactly what time that's my favorite thing about it I always wonder yeah it's just it's a pixel yeah yeah that's really interesting okay any more you think um I don't think there's anything else but I CRM at all um I guess a software company not as important we use custom iio though as our CRM custom iio kind of just anytime someone signs up we send them an email like automations yeah customer iOS it's quite expensive though but I would say like the the best tool by far all of them if I were to rank it is use tol tol is the best affiliate

1:12:39tool and I just think one of the best software tools you can use for any product and then I would say post dog is second best because you get so much data on your product and to make your product better you need all that data and that's just for anyone who doesn't understand just you put one piece of code on your website and you can see where like you said heat map you can see where people are clicking and so wherever their Mouse hovers the most it'll turn like a certain it would literally give you a screen recording of their entire session on do they have to opt into that uh no

1:13:06no that's the coolest part I can also see like literally like their address like I can see everything like it's it is wild stuff how is that legal dude post it's it's just like cookies right it's like all these companies all these companies like have the DAT on you which is and when you see other people data then it's like like you're like holy like other companies have this much data on me on you for everything you're doing on every website this was my major in college and just blows my mind that people don't understand this but yeah it's like how you make really accurate decisions so okay now the last question

1:13:38I always ask people is basically what's like the number one thing that you've learned that you think is true now that you didn't think was true in the past um I think the number one thing that I've learned that's true now is if you kind of just stay consistent at something thing something will end up kind of working I remember in in in middle school I would spend a lot of time just kind of like figuring out what I wanted to do and I I knew I wanted to build something really big I just didn't know what so I started with a clothing brand I started selling chocolates and then in

1:14:07high school I started like a i the first product I did was an info product it was like a Discord for people really yes trading stocks I goes hey $50 a month and and you can be in this Discord and every time the thing that I did kind of got bigger and bigger and sometimes there were fallbacks but I would get back twice as twice as bigger the next time so it's like just consistently doing something uh you you're kind of going to figure it out that's what I've realized like if you just keep kind of going at it no matter what you'll figure it out so this the compounding effect

1:14:34yeah it's just like I think it's more reps I think the saying that I use is like success is 80% consistency and 10% luck and 10% skill like if you're kind of just consistent uh like you're going to get more lucky and you're going to get more skillful and it'll eventually yeah but the biggest thing is just consistency I think that's really important for a lot of people especially young people I think a lot of young people people are like have this like deep like need they need it now so they think that if they're only at 20K bro I like started like I started this when I

1:14:59was 15 if you really think I had to learn all those mistakes uh to get to where I am now that's absolutely crazy okay so I'm introducing a new question I thought from Tom B Bao I think it's like say name Tom Bilu um someone starting into the AI SAS world today what is some what are they going to run into what problems are they going to run into what mistakes are they going to hit if they just start today blind yeah I think if you're not a DB and you have no like thoughts around that then you're going to not know what to like what how your

1:15:25product works and that's like the biggest flaw for most people you need to understand how your product works on like a low level at least like how what's an API what's your going to what's your cost going to be how are you going to get users and if you can kind of understand those you can kind of build on top of that but that's like one of the biggest problems I've seen for even the people that kind of have distribution have a Dev like they actually don't understand their own product like the two other people know more about their product than them and you're supposed to be a PM if you think

1:15:49right so I would say spend a lot more time kind of just understanding like on a low level maybe not like coding but like there's a thing called pseudo code which is just like actually understanding how the code kind of works together and how the product kind of works together and and that that's going to be a lot lot more helpful I think that's going to be a huge problem for a lot of people because people won't do that they'll kind of just try to streamline through and that's going to be the a big flaw once their product has like tens of thousands of users and they

1:16:15actually don't know how their authentication works and why there's a b because they can't actually think of why they kind of just cruise through the steps without actually understanding the first principles of why yeah speed is speed is a it's a double-edged stor like you'll be able to get out there really quickly and that's good but sometimes if you actually don't understand what you're building like to the the te like in terms of low level you're not going to be able to solve the problems that you're going to have as things get bigger and bigger okay last last last final question for real who do who do

1:16:41you learn from um so this this is like really random but every night I spend listening to like a podcast or like a book about someone right now I'm leing Sam wolton founder of Walmart and that book how he scaled up Walmart I took a Southwest flight today and I was like okay I'm going to just look up the founder of Southwest and like his story is insane as well so it's always just like listening to like people that have kind of LED huge teams and done huge things uh and every night I listen to a story and the cool thing about that is like I'll go to sleep and I'll literally

1:17:09imagine that I'm in their position and kind of like carrying out their story like in their eyes and it's super cool you kind of learn from all these mistakes from people that have that made mistakes at the highest level and also had successes at the highest level and it's kind of ingrained into your mind and how to how to make decisions and how to think that might be my favorite thing you said all podcast that's awesome do you follow a specific podcast or you literally seek the big story like big company successes I I I do so in in the daytime I'll listen to like big company

1:17:36stories and like like Sam Walton and like Sam Alman all these big guys right but at night time I'll spend purely on history so like big military leaders um like for example Salah Dean is like a big one for me who was like a military leader for Wow for that dynasty like the these military leaders are also great because they kind of show you how like in their in their time they're like considered Business Leaders right because there was no real business back then there was like only pure war and like how you come to the top of that is like also a great story uh and I I find

1:18:07them more amusing as well so um those are those are some things I listen to okay man I learned so much thank you so much for coming out here and sharing Arab where can they find you one more time uh they can find me on my Instagram at Arab K7 and then on my Twitter Arab K24 all right guys watch this podcast multiple times through because you're going to absorb it the second third time much more and he shared so many little details that are extremely valuable and the most upto-date information in 2024 so shout out to him go follow him on social media everybody have a wonderful

1:18:35day see you next time by