I had zero software uh experience in the past i' done a little bit of coding with python but that was about it um and essentially chat GPT taught me how to build the app and then chat GPT provided the main functionality for the app so it was pretty much entirely built by uh and on chat GPT I'm kind of mindblown by that so wait you built this app that's doing millions of dollars and you just built it by asking chat TP how to do [Music] it welcome to superhuman AI decoding the future where you'll learn what's happening in Ai and how you can use it I'm your host Zan and today I'm joined by Blake Anderson as far as I know Blake's the only entrepreneur in the world that's built multiple apps with chat GPT that are doing millions in Revenue in this conversation I'll ask Blake how he did it and how you can do the same for this episode I'm joined by my co-host husten enjoy Blake thank you so much for coming on the Pod man thank you very much for having me Zane yeah for the listeners I have BL with me here and I also have my co-host and co-founder hust on the podcast as well so we'll just Dive Right In can you Blake can you tell the listeners a little bit about uh yourself and what umax is sure um so I think I can start with the high level where I'm at currently I founded umax umax is a looksmaxing mobile application it enables you to scan your face get ratings and recommendations on how to become more attractive we launched uh start of December of last year so 2023 we're at a little bit over $6 million in ARR before that I built Riz GPT which is now named plug AI that was launched in July of 2023 it's an AI dating assistant that is at two and a half million in ARR um and I'm happy to dive a little bit more into my background from the top if you'd like yeah 100% I'd love to hear more about you know what is it that you were doing before youx yeah I think uh I think the best place to start here is as a child um I've always been fascinated by entrepreneurship and money as a whole at the age of seven I wrote a book titled how to make money as a seven-year-old this was an ebook that was unfortunately never published um but that was kind of the start of my journey throughout childhood childhood I mowed lawns I rake leaves I shovel driveways I sold candy in school I ran bots on RuneScape to generate coins for me which I would then sell on a third party Marketplace I built an ethereum mining machine um I sold counterfeit jerseys on eBay I would pay Instagram meme Pages $2 to promote my jersey um so I guess entrepreneurship has kind of always been in my blood um that then led to college where I honestly spent the majority of my time partying and just hanging out with friends towards the end of college I'd been working on a somewhat failed College Marketplace app incredibly difficult product to scale and I'm graduating and I need to find a way to make money um I had a roommate that was always asking me what he should say to girls on dating apps so I figured that I could just build a chat GPT rapper for it and that's uh that's kind of what led to the Advent of Riz GPT which brings us to where we are today like how do like you start building AI applications so as we all know chat GPT came out end of November 2022 start of December I was taking a course the history of the digital Revolution taught taught by Walter isacson at tane where I attended and what's interesting was I had started playing around with C gpt3 like the gpt3 API just a couple months before chat GPT came out and I was kind of like whoa this is crazy why is nobody talking about this but nobody was talking about it so I figured maybe it's actually not as cool as I think it is but I continued playing around with it then chat GPT comes out and I'm taking this class with Walter Isaacson and something that got me really excited was seeing how excited he was about it this is a man that has an incredibly prolific career documenting uh the stories of the greatest technologist of our time and he his eyes lit up like a kid seeing candy he was like this is incredible and we pretty much spent the entire class just talking about how these new AI Technologies would impact the future of the world at this time I'm working on this College college Marketplace app no AI integration but there was something within me where I was like this is incredible I have to build around this when I was graduating and I knew that I had to make money somehow I was like it's clear that I have to do something in AI um and so that was kind of the origination of my deciding to work in this AI consumer space interesting uh so you said your first thing was called RIS GPT and now it's called plug AI so a lot of folks listening probably haven't heard the word looks maxing or even resz so for for you know folks who are listening can you explain how you found the idea for RIS GPT and then how how did you build it so the idea for it came from a roommate who would ask me what to say to girls on dating apps and you know at the time I'm using chat GPT all the time and I'm like huh I wonder what chat GPT would say to this so we manually typed in the transcript and the response was very AI like but I started thinking about it through prompt engineering and I'm like why why would I not be able to prompt engineer this such that the responses are good and so we started playing around and and ended up doing a pretty good job and this is right at the end of school right as I'm graduating and I'm like I think that there's opportunity here I'm going to build a rapper that helps people with what to say on dating apps at the time I ended up finding out there was an incumbent the Riz app but they hadn't gone viral they they they weren't very popular mainstream um so decided to go ahead and build it I had zero software uh experience in the past I done a little bit of Co with python but that was about it um and essentially chat GPT taught me how to build the app and then chat GPT provided the main functionality for the app so it was pretty much entirely built by uh and on chat GPT Blake I got to interrupt you there because I just want to make sure I get this right so are you telling me that you built this app with chat GPT like chat GPT told you how to build R GPT and that's how it came about to be 100% holy cow in the past people used stack overflow right I used 99% of the code was written by chat GPT I'm kind of mind blown by that so wait you built this app that's doing millions of dollars and you just built it by asking chat TPT how to do it yeah and you know I learned a little bit the first time around but second time around when I built umax maybe not 99% but maybe 85% of the code was still written by chat GPT it took me a couple of weeks but chat GPT took me all the way so we've got the completed app and it comes time to launch and I had brought on two co-founders to work with me on it we were running marketing and nothing was really seeming to stick and then I ended up finding this like untapped Niche on Tik Tok I paid $50 for two separate promos so $100 total and overnight the thing just explodes over 200,000 downloads over the course of the next six days uh $80,000 Mr right off the bat and from there was Off to the Races now mind you we didn't have notifications implemented into the app we didn't request reviews it was the simplest app you could possibly imagine press a button upload your screenshot get your responses that's it um and so from there we were kind of like backfilling all the knowledge that we probably should have learn before launching but I think that this illustrates an important lesson in that you can learn that along the way the most important thing is just like get your product out there especially when it comes to these sort of consumer utility apps now this doesn't apply if you're trying to build a social network right but if you're building a a few function tool there's no reason to delay the launch it's like get it out there get the insights figure out whether or not it's going to work and then you can backfill the things that you need to so then over the course of the next couple of months we scaled up to a little bit over $200,000 an mrr and yeah that's that's kind of the inciting story of uh of Riz GPT now in plug AI That's that's that's amazing right like you know and I think as you said right to summarize like just ship get it out there get the customer insights customer feedback and you can iterate and keep kind of building one thing you mentioned right like so you guys were actually struggling initially to get those downloads so it seems like you you use your hustle you had built up over like several years of Entrepreneurship to again find a niche like what was kind of like the process for you to kind of like come across like like why why Tik Tok why the like you know if someone is kind of thinking about like hey they're in this struggle phase of consumers utility what advice maybe you have for them I think that so I see a lot of one-sized fits-all uh content on Twitter right now telling people or specifically consumer Founders about how to get their product out there you need to use this Tik Tok strategy or this YouTube shorts this Instagram reals this Reddit this x whatever it may be I'm of the opinion that you know maybe there are certain strategies that are working better at certain points in time but far and above the most important factor I think in generating some Buzz virality views is a ruthless cycle of iteration just constant iteration testing out new strategies um RZ gbt took us a couple of weeks to figure it out umax it took us a month and a half to really figure out the winning formula the app that I'm currently working on it's been over a month we still haven't found it granted we're doing well we're at like 100k Mr right now um but I think that there are levels to it right and just because you crossed that first threshold of like generating some views doesn't mean that you figured out your final form you need to keep iterating um I think until until you're finished with the company essentially when it comes to marketing consumer products you just have to always be iterating I couldn't agree more I just want to zoom in and double click a little bit on what you said about building an app at chat GPT because I think a lot of folks who are listening to this are just going to be like wow you can build an app at chat GPT and they're probably wondering how so can you walk me a little bit through the process you said that it took you two weeks can you tell me the process of building an appach chat GPT and how you'd go about it again yeah so I said two weeks it was actually probably more like one month um and I think that the reason that it wasn't quicker was primarily just because I wasn't familiar with the various idees but if I were to go back and do it again um specifically consumer mobile application you start with the designs you need to figure out what you want the product to look like right then you go and you need to figure out where you're going to code it are you going to do react native or you going to do Swift UI or you going to build cotlin as well right these are all important questions from the jump but beyond that getting chat GPT to actually produce the product you just have to be very clear in your language of what you're looking for right so if I have a design with Riz gbt initially it was a gradient background with a logo and two buttons so you just explain that very clearly to chat gbt what's really cool now is that Beyond just explaining you can also upload an image a screenshot of that design and be like hey this is the design I want to build this is the functionality I need for these individual buttons can you please write the code for me so I've built my apps in Swift UI um and so I'm asking chat gbt to wrate the Swift Code I paste that into xcode I preview maybe it didn't get the spacing rate between two bons and then either I'll go in and adjust that myself or I'll ask chat GPT to adjust that um and so I think that the key takeaway here really is that it's a lot easier than you think right you just have to start you just have to ask chat GPT and early on you know there are going to be the The Growing Pains of figuring out how to use the IDE and figuring out why your code won't compile but the the rate of learning is incredible it's like you have your own professional tutor right right at your fingertips so many people are not utilizing it I love it right like you're using you're building apps on top of like GPD and using GPD to kind of like fuel the like an amazing fly wheel you have kind of found uh for yourself so Blake like curiously culy know like you know like are you just like doing this on your own you have some co-founders you build a team and one thing I would love for you like to understand for the like even for the audience is like what's like how do you decide like who to partner with like you know like folks you want to hire bring on like what's kind of like your kind of metrics if only I had the answer to that final question about what really makes a great partner uh I've worked with a number of great people I think that everyone has their own distinct unique uh qualities of greatness I don't have a formula for that yet when it comes to who do I work with um so Riz gbt once again it's now named plug AI I'm going to keep referring to it as RZ GPT uh RZ GPT I brought on two of my co-founders from the college Marketplace app um so it was the three of us after a few months uh one of the other co-founders and I kind of had a difference uh in Creative Vision of where the product should go uh so we split amicably I left I retained My Equity and that's when I started building UMX for umax I was pretty much a solo founder um and then I brought on my older brother he is actually the one that came up with the Madden style rating system which turned out to be the most pivotal de decision um for the the trajectory of the company uh I then brought on an influencer as a co-founder as well so I think it's really situation based in terms of who I choose to work with it's like what's the right person for this role uh what are the gaps here maybe in my skill set that someone else can provide I think that there's a lot of dialogue about how being a solo founder is best in specifically like the Indie hacker Community now that said I like the I think it might be Mark cubin I don't know who said this but 85% of a watermelon is better than 100% of a grape right I think a good uh statistic example here is like imagine you make 80% of decisions correct and another individual makes 80% and there's no overlap there then you're going to make 96% of decisions correct when you combine y'all skill set right and that that can do wonders for the outcome of a company tell me a little bit about you Max I know you've done R GPT you said that you had some differences with the co-founders you want to go do your own thing how do you find the idea for for umax and can you tell everyone a little bit about what the app is and who it's for and what's the problem that it's solving for folks yeah yeah I can start with my leaving R GPT this occurred around middle of end middle to end of October in 2023 and so I left just like that and you know I had income coming from RE GPT but by all intents and purposes I was unemployed right I didn't have anything specific that I was working on at the same time I was uh so I was living at home and I was going through a pretty difficult situation with my family I had been while I was running R GPT I was going to work at Starbucks one or two miles away sometimes my Mom would drive me sometimes I would walk there and back while I'm running this company doing like 200k Mr and so I spent most of November hopping from Friends couches and then uh living on my brother's couch uh all the while I'm ideating and I'd seen looks maxing uh was growing a lot in popularity it kind of seemed to me to be the next step in like the whole lifting social media content at the end of the day the majority of guys work out or lift to appeal more to women um or specifically like Teenage young adult men and looks maxing to me kind of just cut right through that it's like okay let's not say we're just doing this to improve ourselves whatever let's just try to become more attractive and that was really interesting to me so I dove deep into the niche I saw how much it was growing I was like okay I'm going to become a student of the game here I started engaging in all the content I started to try to think like somebody who was interested in look maxing and some of the common problems that I noticed were people wanted to know how attractive they were there were entire subreddits devoted to this people would comment on social media posts can you rate me that sort of thing and then second people wanted personalized tips or so say people wanted tips and this was apparent on social media but the tips provided were not personalized and around this time the GPT Vision API was coming out I was like huh I don't necessarily have to train an ml model in order to analyze an individual's face and provide recommendations I can just prompt engineer this so I started playing around with it and it was doing a surprisingly good job it gave me recommendations to shave my beard when it was a little bit patchy gave me recommendations for skin care when I had some acne on my forehead the light bulb just went off I was like I have to build this so I back at the grindstone again started talking to chat GPT all right round two and kind of within a month of inception I guess a month and a few days we launched um this was two days after Thanksgiving I believe I spent Thanksgiving alone coating I actually had a turkey breast or turkey leg that I cooked myself um and then yeah within the first month so throughout December we did about $100,000 in Revenue January first half of the month we were at around 200k Mr uh slowly growing and then we had a complete copycat like 100% just copied everything that I did and my my plan up to this point was like I wanted to continue growing but I didn't really prioritize trying to go as viral as possible because I wanted to improve the LTV I wanted to improve our offering before B burning the Forest right getting through all of our core users so we had done a couple hundred thousand downloads at this point I was going slow making product improvements what have you competitor comes onto the scene copies our app copies our marketing strategy the whole nine yards um and mind you at this point we're a bootstrap startup right and I'm like okay th this this is a dangerous path that we're entering here either we continue to go slow and maybe they try to overtake us or or they do overtake us or we fight back and the reality is that fighting back meant to some extent um decreasing the amount or the degree to which we could succeed in the long term right because if we prioritize getting millions of downloads when the product isn't in its final form then it'll be tough to reconvert those users but given the state of the market I was like okay we have to do it so within the course of three days I deployed over $200,000 over $200,000 on influencer campaign this was crazy this was like all of our Revenue all of our earnings up to this point deployed within 3 days thankfully it ended up working we went from 200,000 Mr to over 500,000 Mr within the course of like five days uh we did 100,000 downloads one day 990,000 the next day $50,000 in Revenue 40,000 15,000 25,000 um and so that that period was crazy I wasn't sleeping at all I was just 24/7 on the phone with influencers coding working with engineer talking to my social media manager um and yeah since then it's been pretty we've been pretty like we we were growing we hit this crazy Peak and then we've kind of just been uh coting at a little bit over 500k Mr Nice so so so Blake like that that was like a lot of uh great information for our audience and like you know like a crazy Journey so like you kind of have this idea and I I want to summarize this because I think it's really important for audience to like kind of understand like how your journey went you you were struggling like after you left and you kind of ideating an idea like you know like entrepreneurship always has like those challenges and people I think should should understand that and then like once you find the KNE you're actually going deeper like understand and really become like uh someone who understands it like better than 99.9% people you spend the effort figured it out you had a playbook for chat GPD so you're taking the lessons from your past experience and as always right like you you always run into competitors and you did that but you made a bold decision to kind of take the risk of taking all of your revenue and just investing that into so like obviously like it's a very tough decision so like one question here like you know what was going through your mind like hey if this doesn't really work out like what is Plan B that's that's kind of like the first question uh and then the second question is that like as you were kind of like how was you motivating your team through this process I think those are kind of two things I would love to understand um in regards to plan B yeah I uh I'm a very risk-taking individual there is never a plan B it's like either this works or like it goes bad and if it goes bad we'll we'll assess that then um like when I graduated from school no income I don't come from once again I don't come from much money like my brother was giving me loans to pay for groceries uh while I was building R GPT um you're like a true true ENT rur like like really true entrepreneur 100% yeah it's always like doubled down I took the money from R GPT funnel that into umax uh gonna be taking the money from umax funing that into my next big play right um and so yeah no plan B in regards to how do I motivate employees I don't have it 100% solved but I'm a very passionate person as a whole and I think that we should play to who we are and not try to embody somebody else I can believe me I can be very hard at times I can be difficult to work with I can be stubborn as hell um but I always really care about what I do I care about trying and I think that that presents itself to everybody that I work with you said that you're working on something else as well do you want to share a little bit about that or is that a sort of like under the wraps and a secret at this point yeah yeah I'm I'm happy to share about it it's uh so it is called Cal AI it is an AI calorie tracking application um generally I wouldn't share something this early on however uh we're going up against a lot of incumbents we already have like 50 big competitors so I figure that if someone copies us it's like you're not just competing with me you're competing with all these incumbents in addition to the fact that I think that this is the sort of space where like it's so hard to succeed if someone's able to beat us good for them I think that that's highly improbable So Cal AI the thesis here is calorie trackers take a lot of time they're really annoying I personally everyone that I've used I quit after a couple days because it just takes too much goddamn time so take a picture get the calorie analysis in addition to all the other functionalities that current calorie Checkers have you can still scan barcodes you can manually input macros but the key uh differentiating feature here being the ability to take a picture I have two other co-founders on this this is interesting they're both 17 years old we all have equal ownership Zach yary and Henry langmack these kids are absolute Killers they're unbelievable um and so yeah this is kind of my newest Venture I youx is still my main priority um but CI is it's growing quickly we're at about 100K Mr right now um and lot of lot of upside here I saw it uh a tweet that you posted recently uh and you said you know if you weren't logged into gp24 you'd definitely be building on top of gp24 at this point the one that just got launched with you know for listeners who might know it's the the updated version of gpc4 where it comes with audio and assistant like features as well uh so I think the obvious idea if you had to start again or go build something new GPT 40 what are like the sort of problem spaces you would look at or what are some ideas you you try to build I think that there's a lot of opportunity building in the uh audio simulation space right I that people are becoming increasingly antisocial uh and the primary method of socialization right now is like typing right it's it's very anti-human in my my eyes and so I think that building the not just dating simulation but also just like social simulation decreasing this antisocial isation that has occurred over the past decade or two I think that there's a lot of opportunity there um I think learning is something I'm super interested in um so the classic example here is like a tutor a language tutor I've played around with the current GPT for uh you know audio interface in the app right now and it's decent but from what I can tell about gbt 40 it's it's going to blow it out of the water right it's natively uh multimodal I could probably get that functionality just out of the chat gbt app when they implement the 40 voice voice model right I probably don't need a separate app for it I tweeted about this the other day though um that may be the case and I might understand that and I might not go need to use a separate app but the vast majority of people will they won't know that they can that they can leverage that uh sort of functionality within the chat GPT app and so just because you can do something natively within the chat GPT app or within any sort of llm does not mean that there is not a massive amount of opportunity to just package it or reskin it and so one step further in learning recently I've been uh studying complex systems right I like to study uh various Fields just to like try to continue becoming smarter I don't know that the consumer app space is necessarily uh widening my knowledge in deep Tech right or or true academic Pursuits I've been using chat gbt to augment my learning I'm having it quiz me from time to time on where my knowledge is currently at and then provide me resources on the things that I should then go further and learn about right so I'm using this in in the text interface but I was driving the other day and I'm like damn I want to do this right now uh and I tried doing it with the current GPT 4 voice uh architecture and it wasn't great but I was like gp4 would probably be pretty good at this um and so I think setting up a sort of application that enables where I don't have to custom tell it like okay quiz me on this and now provide me with these resources right creating an application that does that natively would be pretty awesome and something that I would use all the time and pay for what is how are you kind of like if you were kind of advising some entrepreneurs who building in the space like what are maybe three things you would tell them to think about as they're trying to build I would say deviate from the norm right I think deviate with what you think what you say and what you do I think specifically in this like SF uh Silicon Valley sort of tech bubble we hear the same things people talk about the same novel Concepts uh or the same Concepts which make them increasingly less novel um and I think it's easy to get caught into the echo chamber but true unique and creative insights come when you pull yourself outside that bubble and you really think for yourself I think that there are a lot of different functionalities that people talk about that are going to be built in the future and as a result you have a disproportionate number of people working on them as opposed to people working on the things that that not many people are talking about and although I don't have a very clear-cut example off the top of my mind um I don't think that anyone was talking about an AI looks maxing app when I built it right and so I think a lot of the alpha Lies when you push past the veil of of the public domain no I totally agree with you there I want to end the conversation on a fun question so again I saw another tweet of yours where you talked about civilization 5 the game the strategy game if anyone's heard of fed uh can you tell me a little bit about why you think playing video games makes you a better entrepreneur I think that video games provide the ability to engage in like an economic or strategic simulation and get repetition across repetition uh rapidly right I can I can simulate uh for example Sims as you mentioned earlier Hassan you can simulate an individual's life in the course of a number of hours right and the various tradeoffs of entering into this career rather than this career they are incredible learning opportunities um to get those repetitions without having to put uh real Capital real time or you know real Stakes on the line yeah 100% I I played a lot of strategy games back in the day more around like Age of Empires and red alert and and things like those but yeah those he knows yeah just of Empires yeah yeah we're we're both raling our age I uh yeah just like two years ago which is disappointingly recent I was addicted to Age of Empires 4 for a little bit once it got released so good so good Blake this has been awesome you're clearly a very smart person but I think you're also a very original Builder I think a lot of folks will look at you know what's trending on the app store or what's going viral on Tik Tok and they'll go ahead and try to build that but it seems like a lot of your building stems from who you are as a person you talked about self-improvement about nutrition uh about exercising and it seems like you really leaned into those and you know all of your creativity stems from your true passion it appears so yeah man thank you so much for taking the time and sharing all this honestly just dope knowledge with us and yeah hopefully we'll talk to you again soon yeah man thank you so much pleasure to meet you guys thank you so much for having me thanks for joining me on this conversation this episode is a part of a short series that I'm recording if you want me to continue these episodes and you want me to record more of them let me know you can drop me a review on Apple podcast or on Spotify podcast or wherever it is that you get your podcast or you can just email me directly at zuph human. that's Z nuph human. thanks for listening again and I'll catch you again next week bye-bye