Latest / The Koerner Office - Business Ideas and Deep Dives with Chris Koerner / Start a Business With Just a Prompt (w/ Bolt CEO)⏐ Ep. #191
Transcript
- 0:00There's this guy CJ on axe, he posted a tweet and said and it
- 0:03cost me 9 bucks on bolt to do this.
- 0:05Then someone asked it how much did you build the client?
- 0:06He was like $9000. Most people still don't know
- 0:11that they can actually build software without being a
- 0:13developer. You literally go from not even
- 0:15having a website to accepting payments in like 5 or 10
- 0:19minutes. Whatever software tool you use,
- 0:21replicate that in Bolt. Worst case you save money for
- 0:23yourself, best case you resell that to others.
- 0:26People will pay so much money for a dashboard.
- 0:28Go start a Dang dashboard agency.
- 0:30What are some prompting best practices?
- 0:32Show us how the average user with no technical knowledge can
- 0:35go from nothing to something on Bolt.
- 0:41This is the greatest time for entrepreneurs to be alive in the
- 0:43history of the world. Why?
- 0:45Because if you're marketing or business heavy, you probably
- 0:47don't know how to code. But now you don't need to know
- 0:50how to code. There are apps like Cursor,
- 0:52Replit, Lovable, Windsurf for sale.
- 0:55So many that are just vibe coding apps use natural
- 0:58language, build a website, build an app.
- 1:01I'm having the best time of my life.
- 1:02And so I'm on a mission to interview the founders of all of
- 1:05these businesses because I want to know from all of them what
- 1:08makes theirs different or unique.
- 1:09And most importantly, how are their users using their app to
- 1:13make money and to save time? What are the common use cases
- 1:17building an agency on top of it, building apps on top of it for
- 1:20others? And what are the fringe weird
- 1:22use cases that people can learn from?
- 1:24So today I had the founder of Bolt dot Nuon.
- 1:27His name is Eric Simons, and he opened the curtain on how people
- 1:30are using Bolt to build big businesses.
- 1:33They went from 0 to $20 million in annual occurring revenue in
- 1:37their first two months because people like me are building
- 1:40really cool things on top of it. They didn't pay me to do this
- 1:43interview. I just think it's a cool app and
- 1:44it's one I use on a daily basis. Hopefully you can learn a lot
- 1:47from it. Please share with a friend.
- 1:48OK, well Eric, why don't you just start by telling us who you
- 1:51are and what you do? Hey awesome, I'm Eric, I'm the
- 1:53Co founder and CEO of bolt dot new.
- 1:56Which is the fastest way to take an idea and turn it into a real
- 2:00web or mobile application? OK.
- 2:02And what makes Bolt dot new different from, you know, other
- 2:05competitors out there? Totally, yeah.
- 2:07I think with Bolt, some things only focus on like web apps.
- 2:10And so we kind of do web and mobile and generally it's
- 2:14insanely fast to get up and running with something that, you
- 2:17know, looks, you know, not just like basic, but really stunning
- 2:20as far as your website or your mobile app.
- 2:23But you know, really importantly, like we really
- 2:24focus on the end to end experience of like actually like
- 2:27building businesses online. I mean, that's not just like,
- 2:30you know, a basic storefront, but having like if you want to
- 2:33build a SAS application or like a new AI application, having
- 2:37payments built in and be able to actually use different APIs and
- 2:40that sort of thing. You know all that is like
- 2:42integrated into this AI agent we've made.
- 2:44Yeah, it feels like a lot of the other Bolts of the world are
- 2:46like, they're really cool party tricks and they're really cool
- 2:49for getting an MVP out there, something that looks flashy.
- 2:52But there's not a lot under the hood either when it comes to
- 2:55design or functionality or accepting payments as you say.
- 2:58Yeah, exactly. And you might be able to do it,
- 3:01but it takes a lot of like of manual work to go and add those
- 3:05things in, right. So you're kind of back at square
- 3:061 then where it's like, well, you're kind of spending maybe as
- 3:10much time then as you would if you had started in an IDE on
- 3:12your computer. Whereas like with bold, it's you
- 3:15can literally go, we can, you know, we can do, you know, a
- 3:17demo of this, but you can literally go from not even
- 3:19having a website to accepting payments on your app in like 5
- 3:24or 10 minutes. And you know, like if and
- 3:26actually building a real application, like a course
- 3:29selling website or whatever have you, you know, and that's crazy.
- 3:32That's never existed before in history to be able to create
- 3:36custom software that you can actually sell like that.
- 3:38What do you think that means for the startup ecosystem like that?
- 3:41It's so easy, so frictionless, so affordable.
- 3:44Like to zoom out a little bit, What does that mean for all
- 3:47these new startups out there? It's an amazing thing, right,
- 3:49'cause the thing that's always been really difficult about
- 3:52entrepreneurship is that, you know, the best entrepreneurs
- 3:56tend to be the ones that have spent a lot of years building
- 3:59companies. And it's largely because, you
- 4:01know, you can't really go and take classes on how to do
- 4:04business and you can, but like, but you don't really learn a lot
- 4:07of practical skills on how to actually go and build products
- 4:10and sell products and everything that goes into starting a
- 4:13startup. And by, you know, being able to
- 4:15like go and take ideas and launch them into reality and get
- 4:18immediate feedback, that shortens the cycle, right, Of
- 4:21you actually be able to try and build a business and really
- 4:25accelerates the learning process for entrepreneurs.
- 4:28If you have to choose a single metric, like what's going to
- 4:30determine if an entrepreneur is going to be successful in
- 4:33starting a company, it's really just a it's a numbers game.
- 4:36You know, if it's how many hits at bat do you get on like
- 4:39building a product, getting feedback, iterating on it,
- 4:42etcetera. If in my view, like this is the
- 4:44tool that my Co founder and I wish we'd had since he and I
- 4:46were 13, learning how to code together through all the things
- 4:49we build over the past two decades.
- 4:51Because it's that, you know, you want to just be able to go and
- 4:54launch things, get feedback as fast as possible and iterate and
- 4:57adapt. So I think we're it's just an
- 4:58explosion of new things that would have never existed before.
- 5:01But I think we're going to see crops of entrepreneurs that
- 5:04that, you know, are able to be successful far faster and and
- 5:09often in many cases where they may not have been able to, you
- 5:12know, afford to just keep taking swings on products.
- 5:15On a scale of 1 to 10, how technical would you say you are?
- 5:18I'm reasonably technical, yeah. I don't know, like 8 or 9 or
- 5:21something like that. So you have like a software
- 5:23engineering background? I do yeah.
- 5:25I never went to college actually, but back so my Co
- 5:27founder and I, he and I grew up down the street from each other
- 5:29in suburb Chicago. And when we were 13, we you
- 5:32learn how to web build web apps together and we've been
- 5:34building, you know, startups and and whatever have you ever
- 5:37since. And he and I have just been
- 5:39best. Friends for the past 20 years.
- 5:41When we went to graduate from high school, you know, we were
- 5:43looking at going to college and it was going to be like 100
- 5:45grand in state for four years. And we're like, this doesn't
- 5:48make any sense. Like we're already making.
- 5:49We can build stuff. What are we doing?
- 5:51Yeah, we were already making like 50 bucks an hour doing
- 5:53website contracting work or whatever.
- 5:55And when you're like 18, that's like meaningful.
- 5:57Six figures. Yeah, exactly.
- 5:59And so yeah, it just did the math, didn't math.
- 6:01And this it was very unpopular back in like 20-10.
- 6:05But I think it's like the world's kind of come around to
- 6:07the idea that, OK, actually maybe forcing everyone to go to
- 6:10college and being 6 figures of that isn't the best idea for for
- 6:14everyone at least, right? So.
- 6:16I applaud you. 2010 is when I was in college and I've always
- 6:18been entrepreneurial. I didn't have to go to college,
- 6:20but I just had this nagging insecurity and maybe a nagging
- 6:23like future dad. I had my first kid in 2010 where
- 6:26it's like this is like a form of insurance.
- 6:28It's a good thing to do. I got 2 college degrees.
- 6:30I've never used either one of them.
- 6:32I feel like AI is taking away all of my excuses, like my macro
- 6:35excuses because when I was in high school to, you know, early
- 6:382000s, I used to say, oh, why wasn't I born 510 years earlier?
- 6:42I could have been in the.com boom.
- 6:43I would have moved out to San Francisco and then crypto comes
- 6:46around and it's like I didn't buy Bitcoin till 2016, which
- 6:49seems early today but seemed very late then.
- 6:51It's like, oh, why wasn't I in the right message board?
- 6:54I could have bought Bitcoin at $0.05, you know?
- 6:57And now with AI, it's like, OK, Chris, where's your excuse?
- 7:00You're on the cutting edge of this.
- 7:02You're on Twitter, you see it like when a new model or new app
- 7:04comes out, you're right there using it.
- 7:06You have no excuse. And the other excuse I used to
- 7:08tell myself is I'm not technical.
- 7:10I can't code. If I could code, I would be, you
- 7:13know, that's my Achilles heel. I, if I could do that and
- 7:15market, you know, if I could be the marketer and the coder, I
- 7:18could build whatever I want. And now it's like I can't for
- 7:21the last like 6 months I've been in these vibe coding amps all
- 7:25day and it is so much more wonderful than I ever imagined.
- 7:28I'm just loving it. I have an idea and I just build
- 7:30it instead of like going to my brother-in-law who's really good
- 7:32at coding. Like hey, can you?
- 7:33Well I'm kind of busy. Like I can just do it myself.
- 7:35It's amazing. Yeah, it's huge.
- 7:37And like, and it's truly like it's the tool that my Co founder
- 7:39and I wish we'd had because like I learned how to code.
- 7:42It was somewhat intrinsically interesting to like learn about
- 7:44programming, but it to me it was always a means to an end.
- 7:47Like I just wanted to build stuff.
- 7:49I wanted to build products, I wanted to build businesses.
- 7:51And, you know, if you're going to, if you're going to build
- 7:54digital based products like you have to, you know, building
- 7:57software is at least was a required part of that.
- 8:00Or knowing how to. Yeah.
- 8:01So I look at today and, you know, some people have this
- 8:04opinion of, oh, well, you know, people should still learn how to
- 8:08code. Like it's, you know, I mean, I'm
- 8:10not, I'm not saying you shouldn't, by the way, but
- 8:12people there's kind of this, Oh, but we're losing something like
- 8:15we also don't speak Latin anymore, right?
- 8:17Yeah. Like things, things change.
- 8:19Like the floor of complexity will kind of go away and
- 8:22meaningfully drop the bar on what it takes to go and build
- 8:25and do things. I think that's a good thing.
- 8:28And anyone that says software engineers are going away,
- 8:30period, like that's just, it's just not true.
- 8:32Like it's right. What's going to happen is you
- 8:33just have developers and this happening already, developers
- 8:37are just focusing on higher value tasks than just like going
- 8:40and changing basic syntax for styles or whatever.
- 8:43And so they're just being way more effective at their job and
- 8:45they are working on things that are very complicated that AI
- 8:48malls are not well suited for. And so it's, you know, maybe a
- 8:51good analogy is like modern industrial manufacturing.
- 8:55It's like there are still maybe some people that say everything
- 8:58should be handmade, every car should be handmade.
- 9:00Like I guess Rolls Royce still exists.
- 9:02They still do that. But it's like this niche market,
- 9:04right? They're expensive.
- 9:06They break down a lot, right? You know, in fact, it actually
- 9:08turns out you having these great, you know, automated
- 9:11manufacturing processes overseen by humans with higher leverage
- 9:15tends to be better for. Everyone holistically, yeah.
- 9:19I mean, there's still people pulling pulling around tourists
- 9:22in horse drawn carriages in Central Park.
- 9:24That doesn't mean that the right decision 100 years ago was to
- 9:26triple down on horse drawn carriages.
- 9:28You buy a Dang car. Exactly.
- 9:31Like I think that big question of do I learn to code or not?
- 9:35It all depends on where you are. Like if I'm 20 something or you
- 9:39know, a teenager like and I'm on the fence about like going hard
- 9:43on business or going hard on coding.
- 9:44Like I'm going hard on business and I'm learning all of these
- 9:47vibe coding tools and that's my coding like my shortcut, right?
- 9:50Whereas if I'm already a coder, like I'm going to learn how to
- 9:53embrace this as much as I can and like you said, delegate and
- 9:56elevate and just start doing more important.
- 9:58Like you'd be dumb as a coder to abandon coding because all this
- 10:02is going to take my job. So let's start looking for like
- 10:04you are in the best position possible right now to further
- 10:07embrace coding, right? Yeah, exactly, exactly.
- 10:10It just gives you so much like understanding how this how
- 10:12software works, how to build software gives you just so much
- 10:15more leverage on how to actually control and use AI tools and
- 10:18agents to to go and do your betting to, you know, change the
- 10:21world or whatever, right. Yeah, it's like the difference
- 10:24between a college like a business professor that like
- 10:26spent 30 years running and starting businesses, like his
- 10:30effectiveness in the classroom versus a business professor
- 10:32that's always only been a business professor, right?
- 10:34Which one do you take? Like I would rather have a coder
- 10:37that knows these AI tools than someone like me, frankly.
- 10:41But I can build these myself and they don't myself.
- 10:43And they don't have to be amazing, they just have to be
- 10:46cool. So I love talking to founders
- 10:49about like the 8020 of use cases of your tools.
- 10:52So I'd love to hear like the 80%, like what are people using
- 10:55it for? What's like the bread and
- 10:56butter? And then what are like those one
- 10:58off use cases that really just make you scratch your head but
- 11:01are really interesting to the listener?
- 11:04Yeah, that's a great question. I mean, this was like kind of
- 11:07crazy things. We launched Bolt like last
- 11:08October and before this we were making a developer ID and so
- 11:12only developers were using our product and the scale of revenue
- 11:15just like we almost went out of business.
- 11:17So like a month after Bolt launch was when we were going to
- 11:19start spinning down the company because we have been around for
- 11:21seven years. We got into the company 700,000
- 11:24of ARR and so Bolt was the last experiment we ran like this in
- 11:27browser development environment technology we made and Bolt, we
- 11:30paired this Frontier AI agent with it that we've made and
- 11:34overnight it just took off like in the first two months we went
- 11:36from zero to 20 million of ARR. Oh my gosh.
- 11:40It's crazy. That's product market fit,
- 11:42right? Like if anyone's wondering, how
- 11:44do I know? Like if you can't sleep because
- 11:47of all the inbound you have, that's product.
- 11:49Market fit exactly and it's the first time I've seen anything
- 11:52like it in my I've been doing startups with my Co founder for
- 11:5420 years. I've never seen anything like
- 11:56this. And because every other launch
- 11:58we'd ever done, it's like you'll have this.
- 12:00If you do great marketing and you'd spend a lot of time on the
- 12:02product, most things you'll have this, you know, tick up on the
- 12:05day of the launch, next day it drops and then it just flattens
- 12:08out back to the grind of OK, we got to make this thing better
- 12:11and that's. OK, that doesn't mean that
- 12:13should quit. Absolutely not right.
- 12:15Like and that is the normal thing.
- 12:16What happened here is from day one, it just the revenue just
- 12:21was cranking and so and our team was super excited on the first
- 12:23day and I was like, guys, temper your enthusiasm like this how it
- 12:26works. Yeah, I've seen this movie.
- 12:27For 20 years, it's coming back down.
- 12:29Yeah. And then the next day it didn't.
- 12:31And I was like, OK, so like maybe the word of mouth, just
- 12:33kind of whatever. And then it just kept going day
- 12:35after day after day. And.
- 12:36And then the other kind of crazy thing was that most of the
- 12:39people that were emailing us as customers for support and stuff
- 12:42were not developers. They were designers or PMS or
- 12:46just people who are entrepreneurs and, you know,
- 12:48people that were salespeople making a website for their
- 12:51daughter. It's like we're like, what is
- 12:53going on? Like this is, this is not the
- 12:55market that we, you know, kind of thought that was going to
- 12:58pick this tool. Up, but it's a much bigger
- 12:59market, right? Massive, right?
- 13:01Massive. And so after the first month, it
- 13:03became very clear to us like, OK, this is something, you know,
- 13:08Bull was really one of the first products that that kind of hit
- 13:10that nerve where it's like, this is not just developers can build
- 13:13software now. And what does that mean?
- 13:15Right. I mean, it's like, dude, there's
- 13:16like a billion people on this planet that now actually can be
- 13:20empowered to go and take their ideas and put them into reality.
- 13:23That's going to change everything.
- 13:24So I'm very enthralled by what you're saying right now when you
- 13:27made that pivot. So from what I understand, you
- 13:29were like a browser extension for developers when you made
- 13:32that pivot, was the reason for the pivot because you were just
- 13:35kind of looking around, seeing that these other guys thinking,
- 13:37Oh my gosh, like we could do that.
- 13:38We're not far from that. They're crushing it.
- 13:41Or was it like some something deeper than that?
- 13:44So I was actually pretty sceptical, I would say of the AI
- 13:47stuff of from 2022 to, you know, beginning of 2024.
- 13:52Not that like Chachi BT was obviously for, I think for
- 13:54everyone in the world was like a mind blowing moment in my head
- 13:57after the first month after that came out, I was like, what else
- 14:01is this beyond like Chachi BT? What's kind of the killer use
- 14:04case here? How is this, you know, one of
- 14:07the next things that this thing's going to get incredibly
- 14:08good at doing that's really valuable.
- 14:10And I mean, honestly, tangibly, there weren't, you know, there
- 14:13weren't a ton of things that that were really progressing.
- 14:16And so kind of by the beginning of 2024, like our company,
- 14:20everyone else in the world was like adding AI features into
- 14:22their app and they didn't make any sense for a lot of these.
- 14:25They're just like AIAI, you know, like a classic hype bubble
- 14:28thing. It was like adding blockchain to
- 14:30your name five years ago. Exactly.
- 14:32And so, so we're looking at it, we're like, we think there might
- 14:34there might be another because back in 21 we the exact same saw
- 14:37the same thing happened with the web 3 stuff.
- 14:39And people were like, you should have web three involved.
- 14:41And it was like, this seems like a bubble.
- 14:43People thought we were nuts. People are like, you should
- 14:45raise money. You should, you know, grow the
- 14:47company to 100 and we're like, no, this seems like a bad idea.
- 14:49Turns out would have been a bad idea, was a bad idea for a lot
- 14:52of people. So I was kind of had the same
- 14:54sort of lens where I was like, I haven't seen anything yet that
- 14:56really strikes me as the breakthrough, like the next
- 15:01breakthrough thing for AI applications that will drive
- 15:04what they're saying is trillions of dollars of potential revenue
- 15:06per year, right? Beginning of 2024, you know, I
- 15:10was using ChatGPT to just generate code snippets and I'd
- 15:12take it and I'd put into our product, our product was an ID
- 15:15for you to build web apps in the browser.
- 15:17And I was like, oh, you know, like ChatGPT should basically
- 15:19have like ARB stuff built into it.
- 15:22So like when you tell it to like build a web app or whatever, it
- 15:24just like does it in the browser.
- 15:26Like that's the magic of we do is like allowing you to run, you
- 15:29know, full, you know, dev environments in a browser.
- 15:32And our vision originally was basically to make building full
- 15:35stack applications as easy as using Canva or Figma.
- 15:38Just you go to a browser URL, just boots, etcetera.
- 15:40Remove the friction. Exactly.
- 15:42And so you know that it kind of clicked.
- 15:44I was like, OK, well, like the code that Chatty BT is writing
- 15:47is like, I think kind of good. So you know, let's make a demo
- 15:50of this. We spent a week doing that in
- 15:51February of 24. Found out opening eyes models
- 15:54just were really not great for code generation like it just did
- 15:57the experience didn't work because you would tell to build
- 15:59something, it would just break. It was ugly.
- 16:01What would come out? Which leaves an opening for you.
- 16:03Exactly what And so and we were like this is a cool idea maybe,
- 16:06but it's not feasible right now with the current state of the
- 16:08frontier models. Maybe they'll improve at some
- 16:10point in the future. Then in May, a couple months
- 16:13after that, we got an early preview of anthropics side of
- 16:163/5 model. It was immediately obvious.
- 16:19I was like, holy hell, like this is huge difference step function
- 16:23difference in in how good this thing is at writing software in
- 16:26the output quality of that. And so we green, you know,
- 16:29brought the project back off the shelf, green lit it and launched
- 16:32it. But so that's kind of the Long
- 16:34story short of it though, is like the best things that I've
- 16:36ever done in my career have always been things where it's
- 16:40been like intrinsically I'm like, this would be kind of
- 16:42cool. Like I am doing, I'm using
- 16:44ChatGPT in this way right now. I wish that I didn't have to
- 16:47like copy and paste to like an environment to run it.
- 16:50Like I maybe we should just kind of build them together And like,
- 16:52I just want that. And like, and our expectations
- 16:55when we launched Bolt is that we were like, this would be
- 16:57incredible if we added 100K of ARR right?
- 17:00By like within a year, we were at 700K at that point of AR
- 17:04after seven years. And so we had no idea how
- 17:07explosive the thing would be. Yeah, it really came from just,
- 17:11you know, my own actual curiosity and and needs versus
- 17:17how to look at the market going, oh, we want to be part of that.
- 17:19Because if you look at like the 2022-2023 era of AI coding
- 17:23stuff, there is tons of people just slapping AI and for that
- 17:28exact reason, because they're all looking around like, oh,
- 17:29they're doing this, they're doing that.
- 17:30But like none of it added value. How did you get the word out
- 17:32when you made this pivot? Good question.
- 17:35We put together a tweet like that was.
- 17:38I was hoping you'd say that. Yeah, you know, it's
- 17:41unconventional, but you know, we like over the past like 7 years,
- 17:45like it's always important. Like when you're building like a
- 17:48startup, you should always be building community around it
- 17:51because you never know which of your iterations is going to be
- 17:54the one that clicks. And ideally you want some amount
- 17:56of built in distribution for the, you know, to propagate what
- 18:00great messages you're going to send out basically, right.
- 18:02So we built up, you know, some like a small following out of
- 18:05maybe 20,000 people were following our account or
- 18:07whatever at the time. And, and, but we, we put out one
- 18:10tweet. Like there wasn't even a blog
- 18:11post, there wasn't a press release.
- 18:13And, and it was a darn good tweet.
- 18:14We spent some time on it And that just basically explained,
- 18:17hey, this is what this thing is. And you know, why it why it
- 18:20exists, why it's different and better than the other stuff
- 18:23that's out there and it went viral.
- 18:25That was important for seeding kind of the initial set of users
- 18:28that came in and tried the thing.
- 18:30But what what kept that that, you know, growth flywheel going
- 18:34was that the experience was amazing, right, like people
- 18:37would, and this is still true today.
- 18:39When you go to bolt dot new, there's no like marketing
- 18:42landing page. You have to sift through.
- 18:43There's no like crap, you know, like I like most products.
- 18:45There's like, oh, this is crap you have to go through to even
- 18:47just try the thing out. I'm not going to lie, when I'm
- 18:49vibe coding stuff I'm looking at bolt dot new for my landing page
- 18:51example. It's like remove the friction,
- 18:54remove the distraction anyway. Totally.
- 18:56Yeah, exactly. It's like just kind of get
- 18:58people straight into the action because that if you have a great
- 19:00product experience, it will speak for itself, right.
- 19:02And so, you know, the prompt box is there, you hit enter within,
- 19:06you know, a minute, you have a web app or whatever that's built
- 19:09that looks great and it's mind blowing.
- 19:11And so people would deploy that URL.
- 19:13They'd be like, wow, I have a website live on the Internet and
- 19:15then they would share that whether, you know, in iMessage
- 19:18or on X or whatever and be like, wow, look who I just made with
- 19:21Bolt dot new. And so that was like, we saw
- 19:23this crazy, you know, organic word of mouth thing happening.
- 19:26And that's like still the primary driver of how people
- 19:28find out about Bolt. Yeah, there's no paid
- 19:30acquisition. Like there's zero $0.00, you
- 19:32know, on the marketing for the launch of the thing.
- 19:35I don't want to go too deep into this but what was the what was
- 19:38the anatomy of that tweet that made it so great in your
- 19:40opinion? Good question.
- 19:43By the way, have you ever heard of the Amazon PRFAQ approach?
- 19:46No. So it's this is a really, this
- 19:49is a good. One, this is good.
- 19:50Yeah. So, so the Amazon PRFAQ is when
- 19:53Amazon wants to go and release a product, before they start
- 19:56making designs or writing code or anything, they write a press
- 20:00release and an FAQ for the product.
- 20:02What are you saying the day that it is launched?
- 20:05Because you effectively have to answer the questions of why is
- 20:08this unique? Where are we innovating on this?
- 20:10What problems is this actually solving right?
- 20:12Proactively removing friction to people ever using your thing.
- 20:16Bingo. And you're starting at the end,
- 20:17right? You're starting at like on
- 20:19launch day and then working backwards from there, right.
- 20:21So and what that provides is then when everyone's built, when
- 20:23you say OK, well, we're going to do this project there.
- 20:26The North Star is so clear on where you actually need to get
- 20:29to right And so for us with the approach we've taken is we have
- 20:32like our approach is a tweet thread FAQ.
- 20:34So what we do is because it's even more distilled format of
- 20:38like A blog post or a press release because you have to
- 20:41every tweet has to have a purpose if it wants to be good.
- 20:44At least you don't have like a run on thread or whatever.
- 20:47And the first tweet, it was really the video demo that we
- 20:49had to showcase like just going to bolt on you typing a prompt
- 20:53and what you could actually build out of it.
- 20:55But one of the guys on our team did a great job with that.
- 20:58As I recall, the text of that tweet kind of started with the
- 21:00classic like, you know, wouldn't it be great if, you know,
- 21:04ChatGPT could actually. I think I'm looking at the tweet
- 21:08right now. It took like maybe a dozen or
- 21:10two of revisions just for the first tweet to get it to where
- 21:13we were, like within 200 characters and etcetera.
- 21:16Was this the one? That is exactly it.
- 21:18All right, why don't you go ahead and read it for those that
- 21:20are only listening? OK, so this tweet says what if
- 21:23AI dev products, Claude V-0, etcetera let you install
- 21:25packages, run backends and edit code introducing bolt dot DM by
- 21:29stack blitz prompt, edit, run and deploy full stack apps, full
- 21:32dev environment MPMV next JS with Frontier AI free
- 21:37exclamation point. That's it.
- 21:38That's the tweet. And then in the comments, then
- 21:40you get to the nitty gritty, right?
- 21:42You don't want to put that up front.
- 21:43Exactly the most. Interested people will start
- 21:45scrolling, but if you put that up front then you just turn
- 21:48people off. And This is why I love that the
- 21:50doing tweet threads is kind of like the focal point of how you
- 21:52make marketing messaging, because the first tweet is the
- 21:55thing that needs to grab people's attention and be like,
- 21:58Hey, this is a thing that sounds dope and is relevant and it's
- 22:02something you would want, right? Like we all agree on that
- 22:04because that's really what you want to get on that.
- 22:06Because once you have them there, they will then read
- 22:08through all of the cool things that this thing can do.
- 22:11It's way easier to make all the the subsequent, you know, kind
- 22:14of tweets by the way, because you're really just speaking to
- 22:16the specific features that really stand out and you know,
- 22:19etcetera. Not that we didn't spend time
- 22:20iterating on those, but that tweet specifically we, what we
- 22:24really want to anchor to is like, you know, the way we find
- 22:26this out is we would just go talk to like our friends and
- 22:28prospective customers or whatever and we'd say, hey, this
- 22:31is what we've built. Here's kind of, you know, the
- 22:33general pitch. What do you think they would ask
- 22:35that question? Well, how is this different from
- 22:37Claude or V-0, right? Why do I need this?
- 22:40Like, what are the actual things that has these?
- 22:42And that's if you look at that tweet, that's exactly, it's
- 22:44like, hey, wouldn't it be great if these things that you know
- 22:47and love could do this? That's what Bolt is you.
- 22:50Lead with a question. Exactly, exactly.
- 22:52So and you kind of have to like you have to talk to people to
- 22:55kind of get a sense of like, what is that?
- 22:56What's the first thing people are going to ask you?
- 22:58Like, yeah, but like, what about blah, blah, blah?
- 23:00You know, it's kind of like the first question on this pocket.
- 23:02What about kind of the other app builders, right?
- 23:04It's like whatever, it's like, well, how do you if we deserve
- 23:07to exist, what exactly is that reason?
- 23:09Right. And like, you know, how are we?
- 23:11That's super important to have. And anyway, so something that
- 23:13we've learned over the years when we've launched up is just
- 23:16really having a very well thought out hook and explanation
- 23:20of what this thing is, why it deserves to exist, why it's
- 23:23going to solve your problems. Way easier said than done to do
- 23:26that well. I think one of the magical
- 23:28things about that tweet was you put the URL right there.
- 23:31Not below the fold, not in a comment, but right up front.
- 23:35And I've looked at the Twitter code and there's a misconception
- 23:38out there says if you put a URL in your tweet, Twitter is going
- 23:41to deboost it because they don't want to take people off the
- 23:43platform, which makes logical sense.
- 23:45So most people, myself included for a while, thought, okay, I'm
- 23:48never going to put links in tweets.
- 23:49I'll put them in the comment tweets, which I would still say
- 23:5280% of the time is best practices, but Elon himself
- 23:56recently came out and tweeted and said, hey, we don't like
- 23:58automatically de prioritize or de boost tweets from the
- 24:01algorithm if they have a link. What we do do is we do whatever
- 24:06we can to keep people in the platform.
- 24:08And so it doesn't have to be a link.
- 24:10It might be, but if your tweet is not interesting enough, maybe
- 24:14because they're clicking on your link and leaving Twitter, maybe
- 24:16it's because it's just not compelling.
- 24:17There's not a good hook, there's not good payoff.
- 24:19We're not going to boost it. But if your tweet is compelling
- 24:22enough and it has a link and people stay on Twitter, like
- 24:24maybe they click the link and they just keep reading all of
- 24:27your comments. They're watching all of your
- 24:29videos. So Twitter's like, wow, I don't
- 24:30give a crap that he put a link in here because they're staying
- 24:33on Twitter even longer than they would have otherwise.
- 24:35Let's boost the heck out of this.
- 24:36Like, sometimes you can really find magic because you lose a
- 24:39ton of people if you put the link in even the first comment,
- 24:42like you lose 80% of people. But if you can find magic and go
- 24:46viral with the URL in the top tweet, oh man, zero to 20
- 24:50million arr in two months. Totally in this.
- 24:53Case totally. Yeah, absolutely, absolutely.
- 24:56And that was the thing too with our the dot new domain because
- 24:58we originally did launch with developers.
- 25:00And so one of the stipulations of having a dot new domain is
- 25:03that when you go to that URL, it has to immediately start some
- 25:07type of action. So if you go to like docs dot
- 25:09new that takes you to a new Google document or a Google like
- 25:12a. Yeah, that's what I used to open
- 25:14all my docs. I never even connected it to.
- 25:16Yep, yeah, exactly. And so so and that's why we
- 25:19chosethe.newdomainandpartiallybecausethe.com was taken.
- 25:21But we were like, we really want people.
- 25:23The magic of this experience is 0 friction between your idea
- 25:27into reality. And so the dot new domain
- 25:30actually requires you to, you know, remove the marketing crap
- 25:33basically, right? I love it.
- 25:36OK, so my ADHD is on full display.
- 25:38Let's get back to the original question that I brought us away
- 25:41from 8020 of how to use Bolt, How people are using it.
- 25:44Most of the people that are coming to Bolt by and large are
- 25:47folks that are entrepreneurs actually like they're building
- 25:50businesses. And then there's the large
- 25:52portion of folks that are, you know, assisting in building
- 25:55businesses like PMS, designers, engineers, even like sales and
- 25:59marketing teams, etcetera. We have two kind of splits of
- 26:02where our revenue comes from, which is 1 is prosumer, which is
- 26:04that's the entrepreneur starting businesses.
- 26:06And then we have B2B, which is existing businesses like Fortune
- 26:105 hundreds or scale ups. And that's where we see a lot of
- 26:13designers, PMS, engineers adopting us to accelerate their
- 26:17prototyping and workflows and that sort of thing, right?
- 26:20The prosumer side and just buy sheer number of heads on the
- 26:23platform. Entrepreneurs is like the number
- 26:25one thing and they're coming here and starting, you know,
- 26:28businesses of like all types, which is insanely cool to see.
- 26:33Do you have some examples like what like selling info products
- 26:37or agency web like they're selling websites or what are
- 26:40they doing? Yeah, totally.
- 26:41So certainly a lot of agencies useful to go and build stuff for
- 26:45their clients, right. That's a pretty large one.
- 26:48And the ROI on that is insane. Like there's this guy CJ on Axe.
- 26:52He, he has the most legendary ROI story we've seen where he
- 26:56posted a tweet and said, Hey, this is, this is the web, the
- 26:58application I made for my client.
- 27:00And it cost me 9 bucks on Bolt to do this.
- 27:03And and then, you know, people got a good amount of pick up and
- 27:05then someone asked the question I was going to ask had they not,
- 27:08which is how much did you build the client?
- 27:10He was like $9000. That's like buying Bitcoin in
- 27:152010. It's crazy, right?
- 27:17If you like search like bold arbitrage, there's kind of a
- 27:19growing number of people that like are kind of connecting the
- 27:21dots on this. There's a lot of agencies that
- 27:23use us for just accelerating the development workflows.
- 27:25And this is becoming a much more, you know, a very common
- 27:28thing at this point that folks are doing.
- 27:30But then you have folks that are starting like SAS businesses,
- 27:33right? There's a lot of different types
- 27:34of these that we see. So you have people that are that
- 27:36will make course building websites for themselves, for
- 27:39example. So instead of going to, you
- 27:40know, one of these other platforms, you can actually own
- 27:43all of your revenue and have your own site to do it etcetera.
- 27:47Instead of giving a 10 or 30 or 50% cut or whatever it is to go
- 27:50and you know, use some other hosted platform and that sort of
- 27:52thing. So that's like just kind of one
- 27:55example of something we've seen other ones too is like people
- 27:57are making AI applications. One example is this guy named
- 28:01Paul who's been part of he was like one of the first couple of
- 28:04weeks of folks that were on our platform, but built an entire
- 28:07CRM with Bolt that has AI integrated into it.
- 28:11And so you can literally industry specific and he was
- 28:13just kind of going after a general this general industry
- 28:16like anyone who's like. A CRM to fit his needs, hoping
- 28:19that maybe other people would like it as well.
- 28:20Yeah, absolutely. And you can go sign up for it,
- 28:23you know, etcetera. We've seen a lot of different,
- 28:25you know, cool types of businesses and folks, you know,
- 28:28making like real money now going and actually selling monthly
- 28:31subscriptions and you know, etcetera for this stuff, right.
- 28:35What kind of app did CJ build the $9000 one?
- 28:38What was that for? That was for one of his clients.
- 28:40It was a dashboard as I recall for some type of reporting in
- 28:44their product, etcetera. And he was, you know, being
- 28:46acting as like an agency, consulting with them and, you
- 28:49know, building out web apps and whatever you're having.
- 28:52That just reminds me of just a framework that I have first
- 28:54principle. Like people love dashboards.
- 28:57I don't care if you own an auto repair shop, you're CEO of a
- 29:00Fortune 500 company. And I think the first principle
- 29:03behind it is like it just makes us feel good that we have
- 29:06everything under control. Like my whole business is on one
- 29:09screen. I know what's going on.
- 29:11Really you don't. But my point is people will pay
- 29:14so much money for a dashboard that is simply just aggregating
- 29:17together. Data via AP is through an
- 29:20interface that you build with something like Bolt.
- 29:22So go start a Ding dashboard agency.
- 29:26That's the thing, you won't be the first person to do it like
- 29:29AI. Total totally and so we're
- 29:31actually running this hackathon right now for the entire month
- 29:33of June where out of you saw this but like so we're I.
- 29:36Did yeah 1,000,000 bucks. The world's largest hackathons,
- 29:39we have over 100,000 people that are participating this at this
- 29:42point, we crossed the world record, which was like 66,000.
- 29:45Guinness is like officially going to market.
- 29:47You know, like I've never done it before.
- 29:48I've never been in the world record, but we're all going to
- 29:50be there. It's been so, so cool to see all
- 29:53these folks going and like, and building ideas that they would
- 29:57not. And the reason we did this, by
- 29:58the way, was that most people still don't know on planet Earth
- 30:01that they can actually build software without being a
- 30:03developer. And so that's like, that was why
- 30:06we kind of created this lightning rod of attention for
- 30:08this hack thumb of like, hey, over $1,000,000 of prizes.
- 30:10You do not need to be technical. In fact, it's better if you're
- 30:14not technical and you come to do this and, you know, come on and,
- 30:17you know, try Bolt for free and like build something and, you
- 30:19know, see what you can build. And there's been amazing stuff
- 30:22for like, folks, there's an app that someone's building in South
- 30:25Africa for giving access to like emergency services down there
- 30:30because apparently there's not really great infrastructure for
- 30:32if you needed to go to the hospital, like where do you
- 30:35actually go? Like it, Which kind of blows my
- 30:37mind because you would expect that this would be kind of a
- 30:39solid problem in 2025. I think it was like a month or
- 30:42two ago, one of his friends had an accident in South Africa and
- 30:46they didn't have Internet access and had no idea where to go.
- 30:50And that's like a common thing. So is there stuff like that
- 30:52where people are building applications that are actually
- 30:56meaningful, kind of even even beyond just a business sense,
- 30:58like actually, you know, obviously important for for
- 31:02humans? I think it's very key that you
- 31:04said like it's actually better if you're not a developer.
- 31:06Like I, I think of this study where they took a bunch of
- 31:09executives and a bunch of kids and they gave all of them a
- 31:11bunch of marshmallows with toothpicks.
- 31:13And they said, all right, build the biggest tower you can.
- 31:16And they put them in groups of like 3.
- 31:17And all the executives, they got together and they had like 2
- 31:20hours to do it. They spent the first hour, hour
- 31:22and a half planning game planning, engineering.
- 31:24Let's get the load bearing right on this.
- 31:26Let's have a big enough foundation.
- 31:27And the kids just started freaking sticking marshmallows
- 31:30with toothpicks and they started building and falling over
- 31:33building involved. And the kids won, right?
- 31:35Because they weren't biased and they just started trying and
- 31:37failing, trying and failing. And when I think of a developer
- 31:40going into something like Bolt, they probably have all these
- 31:43preconceived biases, biases and confirmation biases.
- 31:46And they go into like, well, I don't want to build this because
- 31:48it's probably going to. And they might be wrong in a lot
- 31:50of cases because they have these assumptions that just aren't
- 31:52true. Whereas someone that's totally
- 31:54ignorant like myself can go like, yeah, build me an iPhone
- 31:58app. And then it's like, OK, and
- 31:59you're going to hit problems, you're going to hit friction.
- 32:01It's like, OK, let me take this over to ChatGPT.
- 32:03Let me export my repository to GitHub, OK?
- 32:07And like you just start working through these problems, which
- 32:09that's all entrepreneurship is, is problem solving.
- 32:12And entrepreneurs know how to do that very well.
- 32:14And I can actually see a scenario where the the
- 32:16non-technical vibe coders if you will, are are crushing it.
- 32:21For sure, right. I mean, like, that's 100%
- 32:23because like the big problem with like, you know, the
- 32:27building like apps and software being gated to date on having
- 32:31development experience is that once you're a really proficient
- 32:33developer, you lose touch, you know, naturally with what the
- 32:39experience of using software is as a non developer, right?
- 32:42And, and this is something for us as a company, like we were
- 32:45building tools for developers before, now we are building
- 32:47tools for non developers. And so for us, we are having to
- 32:49completely transform how we approach building, building the
- 32:53experience of our product because a lot of our engineers
- 32:56own instincts no longer actually match with the end users using
- 32:59the product to that. And so who are we hiring to
- 33:01bring in is like folks that are like great PMS and designers
- 33:04that have really good perspective of OK, what really
- 33:07representing the end users experiences.
- 33:09And then for our engineers, there's lots of ways to kind of
- 33:11get that context on the engineers.
- 33:13Like some companies, every engineer has a, you know, knows
- 33:16a customer personally that they talk to every week, right?
- 33:18So like they have that built in like empathy and you know, you
- 33:22know, thread with them, right? But it's a huge advantage for
- 33:26folks that are are building apps is if they don't, if they're not
- 33:29kind of actually bogged down and the bajillion details that go
- 33:34into building a software product.
- 33:35And that's like the stuff that's come out with this hackathon is
- 33:37some of the coolest ideas. And you know, that I have seen
- 33:41from hackathon, largely because it's not just engineers making
- 33:44stuff for engineers. It's stuff that's like, wow,
- 33:47this is like if we brought a girl on our live stream that is
- 33:50a new mom and built an AI app for new moms that helps you
- 33:54like, you know, you can ask it questions about, you know, what
- 33:57do you do? Like how much, you know,
- 33:58generally should you be feeding the baby?
- 33:59You're like, how do you get them on the right sleep schedule,
- 34:01blah, blah, blah. Which, you know, I had, you
- 34:03know, my first kiddo a year and a half ago and I was like, yeah,
- 34:06that's a real problem. Like that's like, you know, a
- 34:09real thing. And, you know, but as an
- 34:11engineer, my first thought wasn't, hey, I should go build
- 34:14an app for this. And, you know, like it just
- 34:16didn't click. That's what's really cool about
- 34:18this, is that it connects people that are directly exposed and
- 34:21feeling the problem to with the tools to actually solve it, you
- 34:24know? Yeah.
- 34:25What are some weird, unexpected ways that people have used both
- 34:28to make a ton of money? Maybe they built an agency on
- 34:30top of it. Maybe it was a random website.
- 34:33Anything. That's a good question.
- 34:34I think the agencies one has been so interesting, but it
- 34:37makes a ton of sense why that's the case.
- 34:40The other really interesting ones that I've seen have really
- 34:44revolved around folks that are going and building, you know,
- 34:49platforms where they are owning the content that they're making
- 34:52and selling it directly and in effect kind of cutting out the
- 34:55middle band of some of these other tools, right?
- 34:57So whether they're selling courses, whether they're
- 34:59selling, you know, materials or access to community or, you
- 35:03know, etcetera, like that's that to me is like some of the most
- 35:06interesting stuff because some of these tools will take a
- 35:09substantial cut out of your earnings, right?
- 35:12And like, yeah, do you before you could have gone and signed
- 35:14up for Stripe and done yourself. But like, you know, the previous
- 35:16company I ran before this was, you know, we were selling
- 35:20courses online kind of like parasite.
- 35:22And for my Co founder, I was just he and him and I, as you
- 35:24know, like a basically a lifestyle business.
- 35:26And it took him and I like a month or two to get just the
- 35:29infrastructure set up with payments and login and da da da
- 35:32before we could even start selling courses.
- 35:34And in retrospect, it probably wasn't worth our time, but we
- 35:38just didn't you know, we're like, we're boots dropping.
- 35:40We need a 30 or 40% cut is brutal.
- 35:43Like we got to pay rent. That to me is, you know, some of
- 35:46the most interesting things from like a monetization standpoint
- 35:49that I've seen for Bolt, but. That could be a good framework
- 35:53of just like whatever software tool you use that is more
- 35:56expensive than you think it needs to be, or takes a bigger
- 35:58percentage cut than you think it needs.
- 36:00Try to replicate that in Bolt. Absolutely.
- 36:02At worst case you save money for yourself, best case you resell
- 36:06that to others. Absolutely.
- 36:07One of my friends, actually a tangible example of this and
- 36:10this, they're not making money on this, but they have there's
- 36:13like a teleprompter iPad app, right?
- 36:15So they're using it to like, you know, they're, they're recording
- 36:16videos and it's like 30 bucks. I don't know of any good
- 36:18teleprompter apps so I would love to hear about this.
- 36:21Yeah, so this blew my mind because he sent me a message.
- 36:23He was like, yeah, like you've been using this app.
- 36:24It's like 30 bucks a month and it sucks on the iPad.
- 36:27And so I went to Bolt and just one shot said make me a
- 36:31teleprompter app. And it did.
- 36:32And it's better. And it, you know, cost
- 36:34basically, you know, two cents or something to me.
- 36:37Yeah. And then they own it forever,
- 36:38right? So there's a lot of things like
- 36:39that where, you know, tools that you're paying for, you actually
- 36:42may not need to pay for them. And in fact, you can customize
- 36:44them for your own use cases for your business, which is like a
- 36:48very common use case actually across the spectrum of people we
- 36:52see, whether it's entrepreneurs or it's actually a Fortune 5
- 36:55hundreds or whatever. Like, they're also using Bolt to
- 36:57just build internal tools for themselves that are kind of
- 36:59bespoke, you know, do. You have any examples of big
- 37:02businesses that have already been built with Bolt, like from
- 37:04an ARR perspective or a number of customers perspective?
- 37:07Yeah, good question. I think I'd have to pull up.
- 37:09I know that there's startups that have launched using US for
- 37:12parts or you know, or full parts of their web apps, whatever have
- 37:16you, because YC startup called Denormalize that used us to
- 37:19build their application and they're doing stuff around AI,
- 37:22data pipelines, etcetera. And they raised a couple
- 37:25$1,000,000 at this point. There's a lot of these though,
- 37:28right? So the one thing is because, you
- 37:29know, folks own their entire Stripe accounts, we don't have
- 37:32observability per SE into how big some of these folks are
- 37:35getting. But we hear this a lot though
- 37:36were they like, hey, we use this to build out this entire thing.
- 37:39And we certainly see the companies that are growing by
- 37:41headcount and adding seats on us, You know, we see that as
- 37:44well. Yeah.
- 37:45I mean, there's a lot of cases where, you know, certainly
- 37:47startups are going to raising, raising money and are making a
- 37:50substantial amount of ARR, but also existing scale UPS and
- 37:52Fortune 500. They're like, you know,
- 37:54launching new products out of this thing that are, you know,
- 37:58doing responsible for 10s of millions of dollars in revenue
- 38:01per year sort of thing. Yeah, I could see someone saying
- 38:04like, well, yeah, I want to get in Y Combinator, but I can't
- 38:06just build something with Bolt to get in.
- 38:08And it's like you can, you can, you can.
- 38:12Yeah, you absolutely can. I'd love for you to show us,
- 38:14like show us how the average user with no technical knowledge
- 38:19can go from nothing to something on Bolt.
- 38:21Cool. Yeah, absolutely.
- 38:22Let me go share my screen here. OK.
- 38:24And for those listening on audio, I'll be describing what's
- 38:27going on here. All right, so we get the Bolt
- 38:29homepage here. Simplest thing in the world,
- 38:31What do you want to build? And there's just a text box.
- 38:32We can tell what we want to build.
- 38:34Maybe let's just have like, you know, create a course selling
- 38:37website, perhaps something like that.
- 38:39So I'm Eric Simons, I want to build a course selling website
- 38:47for my new course on how to use bolt dot new to make production
- 38:54web applications. Anything else you want to add
- 38:57here? No, I'd love to ask you
- 38:59questions as we wait for this to prompt.
- 39:01What are some prompting best practices like the first prompt?
- 39:06Start with like something, you know, relatively simple, like
- 39:09you just start with like the home page and start with maybe
- 39:11some of the, you know, just static parts of the application
- 39:15and like figure out like how the, the colors, the UIUX you
- 39:18want. And that's what this thing's
- 39:19going to go do right now. You want to like, because if you
- 39:21look at it from a typical product development process,
- 39:23that's usually where you want to start.
- 39:24It's just with, OK, let's make sure the general flows, the
- 39:27layout, the look and feel, etcetera.
- 39:29Let's make sure all of that stuff works properly.
- 39:31And then we start like layering on functionality.
- 39:35And so in the next steps here, depending on what comes back,
- 39:37we'll either make some changes to the colors or whatever have
- 39:40you, or to the layout. We're like, OK, cool, this looks
- 39:42great. And let's start adding in some
- 39:44functionality. And at that point you, you kind
- 39:47of want to break these things out, you know, into specific
- 39:51tasks like, you know, prompt by prompt just so that it's easy
- 39:54for you to, if you have to roll back, you're like, OK, this that
- 39:57that didn't really work. There's like only one thing that
- 40:00I was working on there. I can easily roll that back.
- 40:03But generally you don't want to like completely jam a ton of
- 40:05stuff you know, into the chat just because you might end up
- 40:09with results that are hard to rollback and kind of have fine
- 40:12grained control, right? So me, I'm like super impatient
- 40:15and when I'm building something I'm like do this and then it
- 40:17starts. I'm like also do this and then
- 40:19do this and then do this and then do this.
- 40:21Would you not recommend? That it depends.
- 40:23So especially if you're doing like if you're changing kind of
- 40:25parts of the UI and that sort of thing, totally fine.
- 40:29If you're kind of getting into more complex functionality, it's
- 40:31good to like kind of break it down and take it more step by
- 40:34step. I'll show you one feature we
- 40:36have actually that allows you to kind of consolidate changes like
- 40:39that that are maybe large and etcetera.
- 40:41It's called discussion mode. And so this is down here in our
- 40:44chat box is this little like discussion icon.
- 40:47If you click on that, then you can actually just have a
- 40:50conversation where the agent isn't actually building
- 40:52anything. It's just helping effectively
- 40:54answer questions and build a plan on what to go and build
- 40:56with you, right? OK.
- 40:59So this is the app we got back, so it's called Bolt Master.
- 41:02Looks good, gave us some branding.
- 41:03Yeah, like pretty good for. Wow, that looks really good.
- 41:07Jeez. I'm not like a design master,
- 41:09but I think that this looks legitimately good.
- 41:12Let's say for instance it's not good enough for me.
- 41:14I wanna look something, I wanna see something much more sleek.
- 41:17Maybe I have some examples, I can go screenshot it.
- 41:20What prompting tips do you have for asking it to change the user
- 41:24interface to look a certain way? For sure, yeah.
- 41:25I mean, if there's like certain colors or like, hey, I want more
- 41:28animations, you can just tell it to do that.
- 41:30You can just say, hey, I want it to be like a purple theme or I
- 41:34wanted to like. And if you have specific
- 41:35examples, you can actually just take screenshots, drop them into
- 41:38the chat and it'll use that as reference then and go, OK, I
- 41:40will make something that looks more like this, right?
- 41:43How well does it work with screenshots?
- 41:45Let's say I went and took Airbnb's homepage which is known
- 41:48for design. I just screenshot it and said no
- 41:49this isn't vacation rentals but make it look like this.
- 41:53Sure. How well would it do with that?
- 41:54Let's give it a shot. I'll go ahead and drag in the
- 41:57screenshot of Airbnb. Make the theme look more like
- 42:02Airbnb like. Because design is something that
- 42:07is very easy to ignore, but I feel moves the needle much more
- 42:10than people would have said. Absolutely, absolutely right.
- 42:13And so that to me is what's cool is I'm building something from
- 42:15scratch and bolt, what I'm often doing is I will go and find, you
- 42:18know, great design inspiration to pull from and, you know, tell
- 42:22Bolt to to go and look at that and then, you know, spin up
- 42:24design based on that. Right.
- 42:26And hopefully, hopefully the demo gods play well with me
- 42:29here, but it's really looking at the screenshots like.
- 42:33No, no, no. There we go.
- 42:35Go all. What happens?
- 42:37Because this has happened to me in other apps where I just hit
- 42:41brick wall. For instance, I'm trying to
- 42:44build this website right now and it will not let me log out.
- 42:47I click log out, it just refreshes.
- 42:49And I put the code in GitHub and ChatGPT and they all say that,
- 42:52oh, I see what's happening. It's just, it's actually logging
- 42:54you out, but it's not refreshing on the UI over and over and
- 42:57over. And I cannot fix it.
- 42:58Yeah, and it's a very simple thing.
- 43:00What are your tips for getting around that brick wall?
- 43:01Absolutely, because it's going to happen.
- 43:03Yeah, yeah. And this stuff does happen,
- 43:05right. So that's like one of the
- 43:06reasons we rolled out discussion mode is it's very useful for
- 43:09solving these types of issues because it, it's, you know,
- 43:12instead of it typically when you go to other, like even with the,
- 43:16the normal build mode agent in bulk, and this is how, you know,
- 43:19chat P and a lot of the other things by default will work is
- 43:22though, instead of being like, hey, let's like, maybe talk this
- 43:25through and like have you do a couple of things to test stuff,
- 43:27It'll just be like, oh, I definitely know what's going
- 43:29wrong. It's like it has no idea, right?
- 43:31Whereas discussion will say, OK, let's try this.
- 43:33OK, can you give me this information?
- 43:35No, OK. And it'll actually start
- 43:37instrumenting the code base. If things are still going wrong,
- 43:40it'll start adding logs and etcetera to then go in and go,
- 43:43OK, well, what is it? What's happening now, right?
- 43:46That typically goes a very long ways, right?
- 43:49I mean, the cool thing about discussion mode is that every
- 43:52message of discussion mode takes like 110th of the price of a
- 43:56normal agent message, right? And so you can send way more
- 44:00discussion mode messages, which is important for debugging and
- 44:03not just burning through your entire, you know, monthly
- 44:05allotment, right? So this thing, you know,
- 44:07shamelessly, shamelessly. That's pretty close.
- 44:11I mean, there's no shame on this podcast.
- 44:12So we're all about copy paste. Yeah.
- 44:15Yeah, obviously when pulled the exact color scheme and this is
- 44:18like, you know, took exact inspiration from, you know.
- 44:21From those listening, it looks like Airbnb.
- 44:23Yeah, like it took about 90 seconds.
- 44:26And it looks like Airbnb. Airbnb for courses.
- 44:29Yeah. Yeah, exactly.
- 44:31Exactly. I mean, how much time and effort
- 44:33and care and money does Airbnb poured into design?
- 44:36I mean, so like, in all seriousness, why try to become a
- 44:39good designer when you can just do stuff like this?
- 44:41I think for, for a lot of applications, right, I think
- 44:44that you, you don't need to for Bolt.
- 44:46Like we, we spent time on the UIUX, but as I mentioned
- 44:49earlier, like we started Bolt's development in July and we
- 44:52launched in October. So there's 90 days, right.
- 44:54And we poured log into the UIUXA bit to the degree you could.
- 44:58But we didn't even have a mobile responsive view for Bolt and we
- 45:01didn't have it until the beginning of this year.
- 45:02Like we went to zero to 20 million AR without a mobile
- 45:06responsive view, right And I'm not.
- 45:08Saying there's a lesson in there, there is just freaking
- 45:10launch. Absolutely right.
- 45:12If the value you're providing is actually insanely great, which
- 45:15it should be like that's the, that's what you have to do,
- 45:17right? There's all these other things
- 45:19that you should do that people were yelling at.
- 45:20It's like, how do you not have a mobile response to view, you
- 45:22know? Well, we made, we had to cut
- 45:24corners somewhere to get it out online.
- 45:26Yeah. And for a lot of applications,
- 45:28folks build. It's like the design of this,
- 45:30what comes out of these models is excellent and it's why would
- 45:33you go and spend cycles on something that is usually kind
- 45:37of somewhat ancillary to the core value you actually have to
- 45:40go and provide. Not that you should ignore
- 45:42design at all or something, but there's a certain bar you have
- 45:45to hit and unless you're doing something that is designed
- 45:47specific and that's one of the key reasons that your app is
- 45:51going to win and is competitive or something, right?
- 45:54And those are few. Like there's not a lot of those.
- 45:57You should be just be focusing on solving the end user's
- 45:59problems. Do you have any other like
- 46:01prompting hacks like shortcuts like talk to it this way or say
- 46:05this or this phrase or? I'm very curious about that.
- 46:08I encourage people to just like, you know, throw vibes at it,
- 46:12especially for design. And like, you can say, hey,
- 46:13like, go and, you know, make it beautiful.
- 46:16Like make this thing pop, like, you know, add a page that, you
- 46:19know, really focuses on XYZ. Like sometimes it's just better
- 46:22to let the AI take a crack at whatever you have to do versus
- 46:26you trying to like really spec out exactly how it should be.
- 46:29Because I, I think that some people will just get really,
- 46:31really, really detailed. And then what comes out is is,
- 46:34you know, will will like work, but it's, you know, they're
- 46:36spending all this time on things where, you know, in these models
- 46:40that they're trained on, you know, the best, you know, apps
- 46:43in the world. And so you get to kind of
- 46:46leverage a lot of the best practices just by saying, hey,
- 46:48we need a pricing page. Can you just kind of go whip
- 46:50something up with these three prices?
- 46:52And it's like what comes back? It's like, yeah.
- 46:53Oh, OK, yeah, that's great. You know, like I went to thought
- 46:56and it out that way, you know so.
- 46:58In a previous podcast I suggested people do this and I'm
- 47:01curious what your thoughts are on this.
- 47:03Let's say you have an idea for a website and you pull up Bolt
- 47:05alongside four other direct Bolt competitors and you just are
- 47:08prompting all 5 at the same time.
- 47:11Do this. Make this website same prompt in
- 47:13all 5 browser windows. What result are you going to get
- 47:17from Bolt? Either good, better, you know,
- 47:20different, worse, whatever. It doesn't matter.
- 47:22How is Bolt going to be unique when you're prompting 5
- 47:25different, you know, vibe coding apps the same way?
- 47:28Totally. Yeah, I think we've got pretty
- 47:30the best like design output at this point of all the things out
- 47:33there. We've spent a lot of time on
- 47:35that because like again, it's like we're trying to get you
- 47:37from idea to business, you know, as fast as possible, right and
- 47:41in less than 5 minutes. And so if you're having to spend
- 47:44a ton of time on just making the website like look like something
- 47:47that could actually sell a product, that's a problem.
- 47:50And so that's the on the first prompt.
- 47:52That is what you're going to see with bold on the 2nd.
- 47:54Prompt so you'll save time absolutely just on the design.
- 47:56Portal, yeah. And what comes out is just is
- 47:58stunning, right. And and on the second prompt,
- 48:01you're actually, you have a real back end and payments etcetera,
- 48:05integrated with the thing, right.
- 48:07That's what you're gonna really notice.
- 48:08And and the speed of it too, like that's like we really have
- 48:11and this a lot of it comes from the seven years we spent
- 48:13building this technology. Like so this is like how we're
- 48:16running this in browser here. By the way, this is the seven
- 48:20years of technology we spent before bolt on our previous
- 48:23product. And this is no one else has the
- 48:25a pure in browser based dev environment.
- 48:27All the other providers have to use cloud VMS for every time you
- 48:30open up one of their things. So you will notice there's it
- 48:33takes a long time to boot those things often.
- 48:36Sometimes they won't boot at all because you know, the server is
- 48:39borked or whatever, blah, blah, blah.
- 48:40With Bolt, every single time, like when I refresh this chat
- 48:43window, this is booting a fresh operating system on my device
- 48:47inside of my browser, booting it up, installing it, getting it
- 48:49running, etcetera, you know, within seconds.
- 48:51And that's, and you know, on the other guys, this can take
- 48:54minutes to do that. So that's like, I think those
- 48:57are the things that you'll tangibly notice when you kind of
- 48:59go compare these things against each other.
- 49:01How viable is it to build like a production ready iPhone app with
- 49:05Bolt? You can do it, absolutely.
- 49:07Yeah. We have an integration with
- 49:09Expo, which is like a React Native thing.
- 49:11And so on the home screen, you just say like, you know, build
- 49:13me a mobile app, mobile app that looks like Spotify or whatever.
- 49:19This will actually, you know, get a real native application
- 49:23spun up that using just a, you can take a picture of a QR code,
- 49:26you know, that will pop up here, but you can actually run that
- 49:29app on your device and as you are prompting, it'll up like hot
- 49:32reload it, you know, in your hand in real time, right?
- 49:36And so that's like we're one of the few providers that can
- 49:39actually do this. Again, it's because we're
- 49:41running all this stuff locally on device that we can let you
- 49:44build mobile apps and not just kind of like specific types of
- 49:48web apps or something. Yeah.
- 49:50What type of functionalities are still difficult for Bolt?
- 49:53Like let's say I wanted to build a WhatsApp clone and I wanted
- 49:56people to have encrypted chat between, you know, anywhere they
- 50:00are on the earth. Is there any like sort of
- 50:02feature or product or database or something that like Bolt is
- 50:05still kind of young at that that improving?
- 50:08That's a really good question. I think for all of these things,
- 50:10there's kind of a complexity curve of when you start adding
- 50:12in functionality that's reasonably complicated and not
- 50:16super common, you may start. Like an Uber clone.
- 50:20Like an Uber. Clone we're using like GPS data
- 50:22and stuff. Yeah.
- 50:23And especially with Uber, because it's like a lot of what
- 50:25makes Uber work is the dispatching algorithms and
- 50:28software behind. And obviously there's tons of
- 50:30humans that have to support that operation, right.
- 50:32And and that's where things get a lot more complicated because
- 50:35now you're not just talking about, you know, really a like
- 50:38what you know, it's software is like a CRUD based application,
- 50:40create, read, update, destroy where you're just you're
- 50:42putting, you're uploading things to a database, you're
- 50:44downloading things. Now you're dealing with real
- 50:46time systems where you have to have, you know, servers that are
- 50:50doing sophisticated algorithms that you have to invent to, you
- 50:54know, be matchmaking drivers to to passengers, etcetera,
- 50:57etcetera. It's like things of that nature
- 50:58would be very difficult to do with Bolt or any tool for that
- 51:01matter, without having knowledge of it.
- 51:03But it also depends on the use case too.
- 51:05If you're building an app to that, where you only ever have
- 51:0720 drivers to be assigned to different like orders.
- 51:10OK, I mean, you know that that actually would work.
- 51:13Like you don't need some crazy sophisticated, you know, mapping
- 51:17algorithm or something for that, right?
- 51:19And you can just be doing a more direct matchmaking thing.
- 51:22And again, this is like where you're spending a like as people
- 51:24build their projects up, they're spending a lot more time in
- 51:27discussion mode in bold because they're going, OK, I need to go
- 51:30add this functionality, blah, blah.
- 51:31And then and discussion mode will come back and go, OK, we're
- 51:34going to have to go do this. That here's kind of the
- 51:36trade-offs, like what are you actually trying to do here?
- 51:38And you know, really honing in on the plan, which is exactly,
- 51:41you know, when you when engineers write software that,
- 51:44you know, they don't just kind of go, oh, we're building
- 51:46ubergrain and then just start coding.
- 51:48It's like there's you do a lot of planning to kind of really
- 51:51think through, OK, how is this going to work?
- 51:53What's the actual, you know, intended, you know, experience
- 51:55we're delivering here, etcetera, right.
- 51:58And so the same thing is true with this and coming back to
- 52:00earlier in the conversation where it's like knowledge of
- 52:02building software being important.
- 52:04This is why, right? So if you want to build more
- 52:06complicated things, but it's not even like even knowing how to
- 52:09code, it's also just understanding at a high level,
- 52:12OK, what are like the tools and services that that you can use
- 52:16to solve different types of problems?
- 52:17How do you glue these things together, those sorts of things,
- 52:21right? And what we see a lot is people
- 52:24will come here, they will, they'll build their first
- 52:26website, which is something that's maybe, you know, maybe a
- 52:28personal website or it's like a course selling website,
- 52:30something like that. And then they're like, oh, I
- 52:32want to go build something more complicated.
- 52:34And, and then they have to go learn some stuff along the way,
- 52:37which again, discussion mode in our Discord community has helped
- 52:39with, etcetera. You know, folks like, there's
- 52:42something really powerful about being intrinsically motivated to
- 52:46go and learn about this stuff because you're like, oh, I know
- 52:48I can build this. Yeah.
- 52:49And therefore I'm going to go dig in and like, you know,
- 52:52understand how I would do it versus if you don't even know
- 52:54how to code, like it's really hard to kind of get past that
- 52:57point, right? And so here's a mobile app.
- 52:59So this is like previewing it. You saw this QR code.
- 53:01So if you scan this on your device, you can actually, like,
- 53:04run it. Oh, wow.
- 53:05But yeah, So this is Spotify, our Spotify clone and, you know,
- 53:09etcetera. But that's.
- 53:11That's. Zero shot, you know.
- 53:12Yeah, I've never even attempted a mobile app with any of these
- 53:15vibe coding apps, but I want to. Yeah, it's cool.
- 53:17It's fun. I mean, it's, yeah, it's so
- 53:18different to be holding this thing in your hand and then
- 53:21seeing it update live as a native app.
- 53:23It's really it's something else. This was amazing.
- 53:26I know people are going to love it.
- 53:27Where can they find you? We know where to find bolt bolt
- 53:29dot new. What about Eric?
- 53:31Yeah, you can find me on X, I'm Eric Simons over on X.
- 53:34And then our X account for Bolt is Bolt dot new like spelled out
- 53:37DOTBOLDOTNEW. Yeah, that's where you can find
- 53:42us. OK, thanks for watching.
- 53:43I hope you enjoyed it. Share with a friend and if you
- 53:45want to use Bolt, I'd love it if you use my affiliate link in the
- 53:47show notes.