Latest / AI for Founders with Ryan Estes / Drew Falkman builds products at the speed of thought

Transcript

Drew Falkman builds products at the speed of thought · AI for Founders with Ryan Estes

• 0:00
  Welcome back to AI for Founders. I am Ryan Estes.
• 0:02
  Today's guest is Drew Falkman, a veteran product strategist who's
• 0:06
  been building digital products since the dawn of the Internet.
• 0:09
  Drew's career spans 20 years of transforming wild ideas into
• 0:13
  working products for brands like HP, Adobe, Double ARP, and
• 0:17
  American Airlines. He's a former coder turned
• 0:20
  fractional CTO and now one of the leading voices in AI product
• 0:24
  acceleration, helping founders and creators build faster,
• 0:27
  smarter, and with more imagination than ever before.
• 0:31
  In this episode, we dive deep into the new frontier of vibe
• 0:34
  coding, where founders are shipping functional apps in
• 0:37
  hours, not months. You'll learn how Drew uses AI
• 0:40
  tools like Lovable Magic Patterns and Strela to build
• 0:44
  working prototypes faster than Figma, how agentic systems are
• 0:48
  rewarding the rules of product design, and why he believes the
• 0:51
  next billion dollar SAS won't be built, it will be generated.
• 0:55
  If you're a founder, product designer, or creative with more
• 0:58
  ideas than time, this episode is your permission slip to build at
• 1:01
  lightspeed. If today's conversation lights
• 1:04
  you up, go deeper with us at AI for founders.co.
• 1:08
  Every week, we drop stories, tools, and experiments from the
• 1:11
  builders reshaping what's possible with AI.
• 1:14
  Join thousands of founders who believe the next revolution
• 1:17
  won't come from code, it'll come from curiosity.
• 1:20
  This episode is brought to you by Kick Caster, the world's
• 1:23
  leading podcast booking agency for founders.
• 1:25
  At Kick Caster, we connect visionary entrepreneurs with
• 1:28
  podcasts that move the markets. Our clients have landed on top
• 1:32
  shows reaching millions of listeners, raised millions in
• 1:34
  funding, and build brands that audiences actually care about.
• 1:38
  Here's the thing, Podcast guesting isn't just PR.
• 1:41
  It's performance marketing for your personal brand.
• 1:44
  When you Share your story in your own voice, you build trust,
• 1:47
  authority, and influence faster than any ad campaign ever could.
• 1:51
  So if you're ready to turn your conversations into conversions,
• 1:55
  head to kickcaster.com/application and
• 1:58
  apply to be represented today. Kitcaster podcast interviews
• 2:02
  that convert. That's Kitcaster by MO Burst.
• 2:05
  All right, all right, coffee is great, but coffee won't prep
• 2:08
  your meetings for you. It'll just keep you awake while
• 2:10
  you panic, hastily scrolling through LinkedIn at 6:59 AM.
• 2:15
  Ambient, however, will prep your meetings.
• 2:18
  Each morning you get a founder tune briefing sourced overnight
• 2:21
  from your calendar, your meeting history, and the web.
• 2:24
  It's like walking to every call with a CHEAT SHEET that you
• 2:27
  didn't have to write. No more scrambling.
• 2:29
  No more guessing, just context, confidence and focus.
• 2:34
  And if you're raising money, ambient investor dossiers are an
• 2:37
  absolute must have. You'll know the investors
• 2:40
  portfolio, their interests, and their latest moves before you
• 2:43
  even step into the room. You'll even get a heads up on
• 2:46
  red flags. Think of it as the caffeine
• 2:49
  upgrade that your calendar actually needs.
• 2:51
  Prep while you sleep, perform while you're awake.
• 2:54
  Sign up for free at Ambient dot US.
• 2:57
  Go check it out at ambient dot US.
• 2:59
  That's Ambient dot US. Welcome to the podcast, Drew.
• 3:20
  How are you? Great, Ryan.
• 3:23
  It's good to see you. Nice to see you too give us a
• 3:25
  quick introduction to the the projects you're working on.
• 3:29
  I know you have several and kind of like your career with Tae Bo
• 3:33
  on it for us. Yeah, I'll make it fast.
• 3:38
  I've been doing this whole thing for like 20 years.
• 3:42
  I I started out, I studied, I was a liberal arts major and I
• 3:45
  got out of college was like, I don't, I don't know what I'm
• 3:48
  going to do. I was, I said journalism and I
• 3:51
  could see that newspapers were dying and, and, but the whole
• 3:54
  Internet thing was really up and coming and I was like, I want to
• 3:57
  check that out. And I just got addicted.
• 4:00
  I ended up starting a web company in Portland, the couple
• 4:04
  other guys, and we just were building sort of websites for
• 4:09
  all kinds of companies. And I just fell in love with the
• 4:14
  process of figuring out how technology can like enhance a
• 4:20
  company or someone's dream and figuring out what that first
• 4:24
  thing to build is. And so zero to 1 product really
• 4:28
  became my focus. I, I was a coder for a while.
• 4:32
  I still like to code, but that was sort of, I was a, a
• 4:36
  technical CTO and then after that just sort of took off, went
• 4:43
  out on my own, did a lot of work with startups helping them as
• 4:50
  like a fractional CTO, helping them with product strategy.
• 4:54
  Most of the time their CTOS were coders.
• 4:56
  And so they, it's just hard. I've been there.
• 4:58
  When you're CTO and you're coding, it's like thinking
• 5:02
  ahead. When you're just trying to ship
• 5:03
  the next version is, is intense. So I would sort of help them
• 5:07
  with that. And I've done work now for HP
• 5:11
  and Adobe and AARP, American Airlines, like all kinds of
• 5:15
  different companies, just helping to figure out what to
• 5:17
  build and, and getting it built. And lately I've just been, you
• 5:21
  know, I did some Web 3 stuff for a while.
• 5:24
  I have my own startup, but lately I've just, the AI stuff
• 5:27
  especially, it just really fits with the kind of work that I've
• 5:32
  been doing. So I love it.
• 5:34
  Yeah, it's the best. I tell you what, for liberal
• 5:37
  arts guy, me too. And for product guy, me too.
• 5:40
  Boy, it the, the vibe coding kind of, we'll call it a
• 5:44
  phenomenon, is so cool. It's it's so fun.
• 5:48
  And trying to figure out exactly what to do with it right now is
• 5:51
  like my favorite part. So I'd love to hear your take on
• 5:54
  vibe coding, how you're using it, and how you're experimenting
• 5:57
  with it. Yeah, so you know, there's
• 6:05
  there's like a couple levels of it.
• 6:08
  On the one hand, it's really early in the game.
• 6:10
  So in terms of shipping product, it's, you know, a simple things
• 6:17
  are OK, but the more complex they get, the more nervous,
• 6:21
  yeah, let's put it that way. But it's getting better and I'm,
• 6:25
  and I'm watching it and this sort of agentic coding where,
• 6:29
  you know, you kind of vibe code it and then the agents go in and
• 6:33
  kind of clean things up on the back.
• 6:35
  And I've got a lot of hope for that and how that's coming for
• 6:39
  me. I think the best use case right
• 6:42
  now is super rapid prototyping. And you can, because you can
• 6:46
  build a prototype faster than you can in Figma even, or you
• 6:51
  know, these sort of design tools and then it ships as code.
• 6:55
  So it's not only interactive like those Figma prototypes
• 6:59
  where it like you click, there's buttons you can click.
• 7:02
  And that's like how you interact.
• 7:04
  It's actually working. You can fill in forms and it
• 7:06
  validates the forms and like it's an actual app and you can
• 7:10
  get it out there so quickly and you can get it in front of users
• 7:13
  so fast, especially when you combine it with some of these AI
• 7:17
  user research tools. Sure, it's amazing.
• 7:21
  I think I, I was looking at it the other day, you know, there
• 7:24
  for a while I was like, I'm shipping a product a week just
• 7:26
  as a, a practice to see what like I can get done and what get
• 7:30
  finished, you know, and I think I've got like 27 products that
• 7:35
  I've shipped, you know, in varying degrees.
• 7:38
  And, and you're right, the complexity usually off that one
• 7:40
  shot, you're like, oh, this is cool.
• 7:43
  And then you try and get deeper and things start breaking and it
• 7:45
  starts getting weird, you know, so there's, there's definitely a
• 7:50
  handful of projects I have just basically put a pin in and said,
• 7:54
  like, OK, we're going to come back once the kind of like
• 7:56
  cognition, for lack of a better word, is a little bit better
• 8:00
  with the tools. Now you and I know startups are
• 8:03
  built in board rooms. They're built in the wild during
• 8:06
  3:00 AM. Bug fixes, botched launches, and
• 8:09
  that one meeting where everything almost fall apart.
• 8:12
  That's what Code Story is all about.
• 8:13
  It's a podcast where tech founders breakdown the make or
• 8:16
  break moments behind their products, What went wrong, what
• 8:19
  went right, and what nearly wrecked everything.
• 8:22
  Hosted by Noah Labhart, Code Story goes past the highlight
• 8:26
  reel and into the honest stuff, Late pivots, brutal bugs, and
• 8:30
  somehow coming out alive with a product people actually love.
• 8:33
  If you're in the build stage, the scale stage, or the please
• 8:36
  let this feature work stage, I can relate.
• 8:39
  This show will speak to your soul.
• 8:40
  Listen to Code Story at codestory.co or wherever you get
• 8:44
  your podcast. Just search Code Story.
• 8:46
  It'll be right at the top. Because great startups aren't
• 8:48
  born, they're forged one hard moment at a time.
• 8:51
  Code story.co Hey everybody, you know what's crazy?
• 8:56
  I've been sitting on a goldmine this entire time and I didn't
• 8:59
  even know it. My Gmail inbox?
• 9:01
  Literally thousands of warm contacts who already know and
• 9:04
  trust me. But here's the problem every
• 9:06
  founder faces. How do you turn those
• 9:08
  relationships into revenue without feeling like a sleazy
• 9:11
  salesperson? That's exactly why I've been
• 9:13
  using Warmstar dot AI and honestly, it's been a game
• 9:17
  changer for my business. This AI platform connects your
• 9:20
  e-mail, analyzes all your existing contacts, and helps you
• 9:24
  identify who in your network might actually be interested in
• 9:27
  what you're doing. Then it writes personalized
• 9:29
  outreach messages that don't sound robotic or pushy.
• 9:33
  The best part? You review every single message
• 9:35
  before it goes out. No spray and pray nonsense.
• 9:38
  This is relationship based selling done right.
• 9:41
  My listeners know I only recommend tools I actually use,
• 9:44
  and Warm Start has literally helped me reconnect with people
• 9:47
  I'd lost touch with and turn those conversations into real
• 9:50
  business opportunities. If you're a founder struggling
• 9:53
  with outreach, check out Warmstar dot AI and use code
• 9:56
  AIFF to get $50 off your first month.
• 10:01
  Trust me, your existing network is worth way more than you
• 10:03
  think. So some of the things I'm
• 10:05
  building, building like applications for podcasting
• 10:10
  using the camera and the microphone.
• 10:12
  So things like like companions for conversation, I like that
• 10:17
  those are kind of fun things. I'm doing a lot of different
• 10:19
  surveys of walking people down surveys and forking them into
• 10:23
  like different results and then doing really cool reports.
• 10:26
  It's a great way to kind of talk to the the audience,
• 10:30
  particularly with the this podcast, the newsletter
• 10:33
  engagement is just unbelievable. I mean, we have like a $0.50 or
• 10:36
  50% open rate and like 5, 6% click through.
• 10:39
  Like it does really, really well.
• 10:41
  And I think part of that is because I'm always spinning up
• 10:43
  like little surveys and little ways that they can create
• 10:47
  things, you know? So it's like, you know, what are
• 10:49
  the values of your startup? Like having like a little
• 10:52
  application that people can walk through and it kind of spits out
• 10:55
  information for you or prompt generators or I've done places
• 11:01
  you can go eat in Denver, like just all these little
• 11:04
  applications. So when you're building, you
• 11:06
  know, and you're thinking of prototyping, what kind of stuff
• 11:09
  are you doing? What what what is it like Kind
• 11:11
  of captivating your interest right now with some of these
• 11:14
  vibe coding tools? Yeah, it's, you know, it's, it's
• 11:20
  so interesting because you really do, you really can do
• 11:24
  anything at this point. Lately, I've been doing a lot of
• 11:28
  coaching. I I do some coaching for this.
• 11:32
  It's called AI product Accelerator.
• 11:34
  And so I work with a lot of these founders and they have so
• 11:38
  many good ideas. And what's cool is to sort of
• 11:42
  help them get it to the next level.
• 11:45
  And I've seen some things go from like pretty simple apps
• 11:50
  that that are great and they fit a user need and they can like
• 11:53
  dive into the space and plug in AI and make it work right away.
• 11:57
  And then others, like one guy has a, he built this, he used
• 12:02
  cursor. So if you have a little bit of
• 12:05
  programming knowledge, it's crazy how much you can build
• 12:08
  quickly. So he used Cursor, which is a
• 12:10
  little more, it's like an IDE kind of vibe coding duo.
• 12:14
  And he was able to build this app where he could interview.
• 12:19
  If you're interviewing someone in real time, it can flag you
• 12:24
  that this person's probably reading like a response from AI
• 12:28
  because that's what's happening. You know, people are using AI
• 12:31
  and so you can do that and then he's sort of advanced that now
• 12:35
  to like this new sort of AI detection video, AI video
• 12:39
  detection thing. So it can, you know, all this
• 12:42
  compliance stuff and, and, and that's really complicated and he
• 12:45
  built this thing in like 3 weeks is insane.
• 12:49
  But you know, I, I like anything that you can, you know, I use
• 12:53
  lovable a lot. So I think that's fun because
• 12:57
  you can so easily plug in like Chachi BT and stuff.
• 13:00
  So you can take something and build like one of the things
• 13:06
  that I've I've built is this travel app.
• 13:08
  So you can build an itinerary in the travel app and it's so easy
• 13:14
  and go out and say, I want to go here and the kind of traveling I
• 13:18
  like to do, I like to do outdoor, I like to do, you know,
• 13:23
  fine dining, whatever. And then it'll like generate an
• 13:26
  itinerary for you and a schedule and then you can book in the
• 13:31
  linking sites. And, and that's where it gets a
• 13:33
  little complicated. So I tend to just like, do the
• 13:35
  first version and put it up and then let people use it, you
• 13:38
  know? Yeah, totally.
• 13:40
  And see if they like it and then if they do and they're using it
• 13:43
  get gets viral a little bit and so you can then you put a back
• 13:46
  end on it. Then you figure out ways to
• 13:48
  monetize it. Like this is the path, you know,
• 13:50
  it's like build the audience that trusts you and is looking
• 13:53
  for what you're looking for or is looking for what you're
• 13:55
  putting out there and then test a product with them totally for
• 13:58
  free and then build around the community that grows to love it.
• 14:01
  I just love this whole model. You know, the a buddy of mine
• 14:06
  made postmaloneyourself.com and you just upload a picture and
• 14:10
  you there's a threshold of 0 to absolute posty where you can,
• 14:15
  it'll like update, you know, your picture, then you can post
• 14:18
  Malone yourself. So like, these are the kind of
• 14:20
  products that I'm really interested in, which are kind of
• 14:23
  novelties, you know? I mean, it's, yeah, I'm, I'm
• 14:28
  working on doing stuff. So I've been looking at
• 14:30
  platforms because I do all these courses and I've done a lot of
• 14:33
  stuff with LinkedIn learning, but I kind of want to do more of
• 14:35
  my own. And I think you can be so much
• 14:39
  more nimble. And so that's kind of my next
• 14:42
  thing I want to work on is building something where people
• 14:45
  can come and learn and I can help figure out what people want
• 14:49
  to learn, right? And sort of guide them down that
• 14:53
  path. And you know, I don't want to
• 14:55
  make it to AI automated. That's the other thing.
• 14:59
  But there's so many opportunities, like you said,
• 15:02
  when you have a good audience and you understand what they
• 15:05
  need, you can build something so quickly, more quick than ever
• 15:11
  before. Like if I was a SAS company, I
• 15:14
  would be really, really worried right now because it's so easy
• 15:19
  to just build your own. Like if I'm going to look at it
• 15:21
  as an e-learning platform or I can just vibe code my own,
• 15:25
  that's exactly what I need, right?
• 15:27
  Because you think of most SAS, there's like 80% of the stuff is
• 15:30
  for users that are not you and you use like 20% of the stuff
• 15:36
  and they're always doing upgrades to all the things.
• 15:38
  And sometimes it's yours, but a lot of times it's not, right?
• 15:41
  And there are other customers that are using that 80%, but not
• 15:45
  using the 20%. So why not just build that 20%
• 15:48
  that you need, right, and be modifying it yourself and build
• 15:52
  everything for yourself. And honestly, I think that's
• 15:54
  what's that's what the future is going to be.
• 15:56
  And maybe there's a place for SAS to build sort of an API back
• 16:00
  end to, you know, plug in some things, but it's going to be
• 16:04
  really different in five years. You know, if I was a SAS
• 16:07
  company, I would be thinking very hard about what that might
• 16:10
  look like. Yeah, and all kinds of different
• 16:12
  pricing models have emerged, whereas use bases or outcome
• 16:16
  bases, It's really interesting. You know, one thing I'm really
• 16:19
  looking at too is yesterday was that Stripe released their
• 16:24
  agentic commerce protocol with GPT.
• 16:27
  People can learn about this at chatgpt.com/merchants, where
• 16:31
  essentially what they're rolling out is that within GPT?
• 16:34
  So if I'm like, hey, you know what time it is, where what are
• 16:37
  the best black sweatpants, you know, and it's going to give me
• 16:41
  some links. And now the the vendors would be
• 16:44
  able to sell inside the GPT chat.
• 16:46
  So it'll send you I think it's, it's rolling out with like Etsy
• 16:50
  and I think it's rolling out with like the top 1,000,000
• 16:52
  Shopify stores. So independent merchants can now
• 16:56
  sell directly from people's chat in GPT.
• 16:59
  But what's different from like Amazon or Google is that like
• 17:04
  the merchant gets to maintain the customer relationship.
• 17:07
  So the merchant gets the e-mail address, they get all the same
• 17:11
  information as if they are purchasing straight through
• 17:13
  their Shopify store. They'll be able to keep that
• 17:16
  with GPT. Now there's 50 billion product
• 17:19
  searches like I think on a monthly basis through GPT.
• 17:23
  So if you're, you know, you're the little mom and pop shops
• 17:26
  been getting your butt kicked on Amazon or you're, you're kind of
• 17:29
  bullied into like having to spend more and more money to
• 17:33
  advertise on Amazon and Google. Like now you know, if you're
• 17:36
  optimizing correctly for chat or where, what are they calling it
• 17:38
  Geo? Now you have an opportunity not
• 17:41
  only to sell, but then you can like grow the customer
• 17:43
  relationship beyond that initial sale, which I think is
• 17:46
  absolutely unbelievably cool. And hopefully like this kind of
• 17:50
  becomes a protocol for LLMS in general is like the merchants
• 17:55
  are able to retain some some of the customer data and the
• 17:58
  relationship. Yeah, yeah.
• 18:00
  Cuz I think that's the challenge now with when you're trying to
• 18:04
  discover things, it's, it's become using any sort of GBT is
• 18:09
  you have the, the discovery process and then you have the
• 18:13
  purchase process. And so for a lot of companies,
• 18:16
  they're trying to become discoverable, but then you have
• 18:19
  to somehow get to that conversion.
• 18:22
  So it's like that, you know, what's the best, you know,
• 18:25
  whatever, what's the best car rental company, we're going to
• 18:27
  get the best deal. So you, they might come up, but
• 18:30
  then it's not usually a click through link, right?
• 18:33
  Like it was with Google search. So it's really changing the
• 18:36
  whole paradigm of how companies can get in front of people with
• 18:39
  their products. That's cool that stripes doing
• 18:41
  that, but that's going to that's going to be a huge place.
• 18:44
  I mean, it has to happen, you know, eventually because it's
• 18:48
  people are going to, people are already starting to use this
• 18:51
  more than Google, you know, even the Google overview is, is
• 18:56
  cannibalizing their regular search, you know.
• 18:59
  Well, and we'll have a genetic shopping, you know, so that
• 19:02
  basically, you know, I had a conversation, great one with
• 19:05
  Crossman last week and it was talking about, you know, how
• 19:08
  we're going to be able to do commerce through, through AI,
• 19:15
  which it has to be some kind of like stable coin.
• 19:18
  It has to be on the blockchain because nobody's given their
• 19:21
  bank accounts to some AI agent. But if you have like a OK, I
• 19:25
  have 100 bucks in wash in my browser.
• 19:27
  And if I'm just like, yeah, give me those sweatpants, boom, boom,
• 19:30
  boom, boom, done. You know, it's just like taking
• 19:32
  care of it, you know, so you're replenishing like this.
• 19:35
  The, the, the agents bank account and then now the agents
• 19:38
  are paying other agents and all, I mean, it's about to get really
• 19:42
  cool and really fast, you know, removing all the friction,
• 19:45
  removing all the two factor authentication, which man, I, it
• 19:49
  sounds like first world problems, but it's so irritating
• 19:52
  to me. Like, you know, give me, I'm
• 19:54
  texting you a 6 digit number. All that stuff is going away.
• 19:58
  So you know, what I'm really interested in is like, hey,
• 20:01
  what's 3 four steps down the road?
• 20:03
  You know, how are we getting to like interface with computers
• 20:06
  and with AI? That's not just like glowing
• 20:08
  rectangles, rather it's an earpiece or maybe it's in your
• 20:11
  glasses or maybe it's just some kind of pulse on a ring that's
• 20:14
  giving you information and you're triggering it with by
• 20:16
  wiggling your ears or something. Like we're going to get into
• 20:19
  this world of predictive AI and, and kind of human computer
• 20:24
  interface that's like not. Like what it is now.
• 20:28
  So there we're in this fun intermediary step, which is
• 20:32
  really where I'm looking for like all these companies and all
• 20:34
  these like kind of stepping stones to the next thing.
• 20:38
  The fun part too, is that the people that are building the
• 20:41
  stepping stones don't know where it's going.
• 20:44
  You know, it's kind of hashing out as we go.
• 20:47
  Now you're coaching so good candidate for for coaching for
• 20:52
  you. Is it is it a an idea guy that's
• 20:55
  wondering how to use kind of some vibe coding skills?
• 20:58
  Is it is it somebody who's been a long time coder?
• 21:00
  It's like trying to figure out how to get the most out of vibe
• 21:03
  coding tools. What?
• 21:04
  What kind of courses are you teaching?
• 21:06
  Yeah, most of the work I do are, is with product managers and
• 21:11
  designers or sometimes founders as well, where they, you know,
• 21:16
  people who are not as technical and, but they have ideas and
• 21:21
  they, and for the first time, I think they can actually make
• 21:25
  that connection without having a design team.
• 21:28
  And like a, they don't have to do all that stuff.
• 21:30
  It's like you think it through. And so I just helped them to be
• 21:33
  like, OK, this is what you need to plan for vibe coding, because
• 21:36
  you do still need to plan, you know, and then this is what,
• 21:39
  once you've got a plan, this is how you build it.
• 21:42
  And there are some technical things you need to know to, to
• 21:45
  do things right. You need to have a basic
• 21:47
  understanding of like what a database is SQL, like even
• 21:51
  lovable, which to me is one of the most friendly sort of vibe
• 21:55
  coding tools. It'll say like, I'm going to
• 21:58
  create this database table for you And, and it's like a sequel
• 22:01
  dump, right? And you look at it and you're
• 22:03
  like, OK, you know, if you don't know sequel that it's like, OK,
• 22:08
  I hope that's right. So I really helped those kind of
• 22:10
  people to to get things up quickly.
• 22:13
  Yeah, let's talk about some tools.
• 22:15
  So the the the vibe coding, you walk into it, you prompt and
• 22:19
  make something, you're like, Oh my God, I'm a genius.
• 22:21
  You know, whether that's Replit, if that's Bolt or if that's pod
• 22:25
  code or it's lovable, all these great platforms.
• 22:29
  One kind of complaint I have, and it's not really a complaint,
• 22:32
  but like once you build the tool, you're sitting there and
• 22:34
  it's like, boy, this looks very vibey, you know?
• 22:36
  So I name everything vibe this vibe that not trying to hide it,
• 22:39
  but I would like to get make it more beautiful.
• 22:42
  Do you have good like ways to design the UIUX so it doesn't
• 22:47
  look so kind of blocky? Vibe code basic Y.
• 22:52
  Yeah, I mean, it's getting a lot better.
• 22:54
  Some of these tools are are are better.
• 22:59
  You know, there's, there are paths from Figma to things.
• 23:02
  So if you're a designer or if you know a designer, I mean,
• 23:06
  Figma has kind of become the standard of how you design UIS.
• 23:11
  And so a lot of them will support like you can fully
• 23:15
  import builder IO is sort of a, they were an intermediary tool
• 23:20
  where they would, you could pull things in from Figma and then
• 23:24
  you could pull it out to somewhere else.
• 23:25
  And they, they're kind of building their own vibe coding
• 23:27
  system now, which is getting better.
• 23:31
  And I've used their tool that you go into their system and
• 23:34
  then it goes into lovable and that that works pretty good.
• 23:38
  One of my favorite tools now is, is magic patterns.
• 23:42
  So magic patterns is it's, it's like those tools, but it only
• 23:47
  does the front end. So they basically said, let's
• 23:50
  just focus on the front end and build that out.
• 23:53
  And you can actually import directly from Figma.
• 23:56
  So you can just pull in, you like get the layer ID and it
• 24:00
  pulls it in. You can upload screenshots.
• 24:03
  So you can, you know, get a screenshot of whatever it is you
• 24:07
  want to build and then you can start from there.
• 24:10
  So if you're an existing company and you want to like test out a
• 24:12
  new feature, you can do that. And then you can have your own
• 24:17
  sort of design systems effectively built in that
• 24:20
  include components and all your styling.
• 24:24
  And, and it can be you can say like I want to clean and light
• 24:27
  and modern, or you can be like, you know, all corner radiuses
• 24:31
  should be 6 pixels. And I want to have, you know,
• 24:34
  these are my primary colors and these are my secondary colors
• 24:37
  and it'll handle it. And then again, you have all the
• 24:39
  components. So anytime it puts a button,
• 24:41
  it's going to pull in the button component.
• 24:44
  And you can have different things of it.
• 24:46
  And you can create them all in vibe coding or you can sort of
• 24:49
  plug in code for them or import them from Figma.
• 24:53
  And once you get there, it's so fast and powerful.
• 24:56
  So it's really good for like testing new features in existing
• 24:59
  companies, building prototypes for validations.
• 25:02
  That that's my favorite use case, honestly.
• 25:04
  You take magic patterns and you build a prototype and you can do
• 25:08
  this in like a couple days. And then there's this other tool
• 25:12
  called Strela and they're going to have a self-service version
• 25:16
  coming out soon. And you basically it's a, it's a
• 25:20
  tool where you say, I want to ask questions and it'll, it'll
• 25:25
  recruit people and then they meet online with the AI and it
• 25:29
  has a friendly voice. And it's like a little sort of
• 25:32
  vocal vocalization thing. And you chat with it and you can
• 25:37
  have it, it can open up a URL and it'll record everything
• 25:41
  you're doing and you talk about what you're doing and it'll ask
• 25:43
  you like dynamic follow up questions based on what you're
• 25:47
  responding. And then when you're done with
• 25:49
  all that, it does all the analysis.
• 25:52
  So you literally just like spend a couple days building a
• 25:55
  prototype, You build this questionnaire thing that it, by
• 25:59
  the way, there's like AI tools to make that go really fast.
• 26:02
  And then you, it recruits a bunch of people.
• 26:04
  So like in a week you can get really decent validation of user
• 26:09
  prototypes or feedback, you know, However, whatever phase
• 26:13
  you're out of it, that kind of stuff just is insane.
• 26:16
  That used to take like we would do consulting engagement for
• 26:19
  this and it could take 2-3 months to do that process.
• 26:23
  No, it's so exciting that strela.ioandmagicpatterns.com
• 26:29
  those tools are awesome. I've never heard of those.
• 26:31
  Thanks, Drew. Really good I I highly recommend
• 26:34
  them. Yeah, but I, I kind of left for
• 26:36
  a while shipping products and doing apps and tools and stuff
• 26:40
  for entertainment stuff. And I, the first one I did, I
• 26:44
  did with Fernanda and we recorded it and we were just
• 26:47
  trying to make a, a video pat patching together some nano
• 26:50
  banana stuff at with the script. So we went all the way through a
• 26:53
  script and blah, blah, blah, blah, blah, put it out it it
• 26:55
  looked pretty horrific for a first draft.
• 26:59
  But obviously we're kind of doing it fast and doing it live.
• 27:02
  But I, I've been developing that.
• 27:04
  So I'm very interested in like, you know, creating repeatable
• 27:09
  scenes, creating repeatable characters, you know, basically
• 27:12
  using whether it's mid journey tied to GBT, tied to NA Dan,
• 27:20
  basically automating the whole all the prompting systems until
• 27:23
  it gets good enough where you can, you can use multiple camera
• 27:26
  angles, you can create environments, you can create the
• 27:29
  filters, you have cohesion and you can start to make movies,
• 27:33
  you know, And so had a little bit of success doing like 62nd,
• 27:38
  92nd movies. I want to give a shout out to my
• 27:40
  buddy Danny Newman, who's doing some really cool stuff too.
• 27:43
  It's not quite there yet, but as far as like the generative
• 27:46
  video, audio, images, do you have any recommendations for
• 27:53
  those? I, I don't really, honestly, I,
• 27:57
  those tools are amazing for me. I'm talking to my son about that
• 28:01
  because he's really into like screen writing and stuff and
• 28:05
  just telling him you got to do this.
• 28:07
  Is it? Are you finding that there's
• 28:09
  character consistencies? Because some of the stuff I've
• 28:11
  seen is you do some different scenes and the characters kind
• 28:14
  of there's that continuity of the characters.
• 28:17
  They look a little different in place to place.
• 28:20
  That's the challenge. That's the challenge for sure.
• 28:22
  I was using Danny's tool and it's very cool.
• 28:25
  There's a story I wrote about my daughter and she turns into a
• 28:28
  Mountain Lion. It's like a kid story and
• 28:31
  there's this great transformation that it nails it.
• 28:33
  But you, you know, the, the character's a little bit
• 28:35
  different that you know what I mean?
• 28:36
  In every single scene. And it in the like final scene,
• 28:39
  the, the Mountain Lion was actually a regular lion.
• 28:42
  So it looks like Pixar, it looks great, like this character
• 28:47
  consistency thing. This is a problem with AI, you
• 28:50
  know, that they've really known how to to overcome this yet,
• 28:53
  because it's not like they're we always think of or I think of
• 28:57
  like, you know, creating content like how they used to make a
• 29:01
  cartoon is they'd put a background and then they do
• 29:03
  cells over it in the background stays and they move and they
• 29:06
  make cells. But it's that's not how it is.
• 29:08
  Every scene is completely generated fresh every single
• 29:11
  time. So like to find the nuance and
• 29:13
  all the different physics of every single angle and
• 29:16
  everything. It's really, really hard, you
• 29:19
  know, but they'll fit, they'll fit fix that eventually.
• 29:22
  So what I've been thinking about is how do you, how do you strip
• 29:25
  that away? You know what I mean?
• 29:27
  Like you go full Pixar or let's say you just go full like
• 29:30
  cinematic. This is going to be very
• 29:32
  difficult because the rustling of the trees can change.
• 29:35
  But what if you did something else?
• 29:36
  What if you go super minimalistic and it's like some
• 29:39
  Japanese wood Brit or wood print type stylistically where you
• 29:43
  could have more control, foreground, background and have
• 29:46
  things kind of emerge. So I've been playing with that
• 29:49
  and finding like very universal, heartfelt stories that people
• 29:53
  would be really into, that you can draw them into the story and
• 29:56
  personalize it. I see this, the world part of
• 29:59
  what the where the world is going is like super specific
• 30:04
  bubbles of one person and what they like.
• 30:07
  So if the new Superman movie comes out for me, you know, it
• 30:11
  might be George Clooney. For you, it might be Dolph
• 30:14
  Lundgren. I don't know.
• 30:16
  Why him? Maybe you like Drago.
• 30:17
  I'm I'm assuming you do. They never doubt.
• 30:21
  Or for somebody like you, you'll be able to personalize
• 30:23
  everything. It it'll, it's going to know
• 30:25
  what kind of ending you like. It's going to it, it, there's
• 30:27
  going to be like, there's going to be like a, a choose your own
• 30:31
  adventure type situation for a director, where, what are they
• 30:34
  doing? They're like creating
• 30:35
  opportunities for you to be entertained.
• 30:37
  And it's, it will kind of run itself, you know, and then you,
• 30:41
  it could be even, you go out a little further.
• 30:43
  It's like checking your Apple Watch for your pulse and your
• 30:46
  sweat and your pheromones. So as you're reacting to the
• 30:49
  movie, it's going to start to guide you in certain directions
• 30:52
  based on like what it determines to think you're going to like
• 30:55
  anyway, that's the way down the the road.
• 30:57
  But like this kind of personalized entertainment is
• 31:00
  going to become more and more and more relevant until people
• 31:03
  are exchanging what's most beautiful and valuable to them.
• 31:06
  And out of that collective, if you have these cultural kind of
• 31:09
  like artistic tides, it's going to be unbelievable because it'll
• 31:13
  be so unique and so undeniably awesome.
• 31:16
  You know, problem with the with the movie industry right now is
• 31:19
  like they're putting out garbage movies and nobody wants to go to
• 31:22
  the movies anymore. I know it's so I was just
• 31:25
  complaining because there's like another Tron out and it's like
• 31:29
  Tron's OK, but why not just make a new sci-fi movie like in it's
• 31:33
  not even like in the same theme. It's like all takes place in the
• 31:36
  real world, not inside of a sort of logic board.
• 31:39
  So it's, yeah, it's just they're just rehashing the same stuff
• 31:43
  and all the superhero movies and all that, and it's, it's, it's
• 31:46
  got to change. I agree totally.
• 31:49
  I mean they films are expensive, I get it.
• 31:51
  But imagine when there's a billion film makers and the best
• 31:54
  ones just emerge and be like, whoa, Dang, look at that.
• 31:57
  It's going to be some really cool stuff.
• 31:59
  We got a couple more minutes here.
• 32:01
  Are you ready for the Lightning Round I?
• 32:05
  Don't know baby, I'll give it a shot.
• 32:09
  As ready as I'm going to be. See.
• 32:10
  What happens? No.
• 32:12
  Cool. This episode is brought to you
• 32:17
  by MO Burst, the AI first growth engine trusted by iconic
• 32:20
  startups like Calm, Robin Hood, Reddit, Uber and Google.
• 32:26
  Moburst is a digital, full service, mobile first marketing
• 32:29
  agency that helps companies to scale and become category
• 32:32
  leaders. Whether you're finding product
• 32:34
  market fit or ready to accelerate growth, Moburst
• 32:37
  delivers creative, data-driven solutions that help with scale
• 32:41
  and efficiency and impact. At Moburst, we're not just
• 32:45
  guessing what works. We're predicting it.
• 32:47
  As an AI first agency, Moburst uses cutting edge tools to
• 32:51
  analyze user behavior, adapt creative in real time, and cover
• 32:56
  hidden growth opportunities. The result?
• 32:58
  Marketing that's faster, smarter, and built to convert
• 33:02
  from early stage startups to global brands.
• 33:04
  We've helped hundreds of companies overcome complex
• 33:07
  challenges, unlock new audiences, and optimize every
• 33:11
  metric that matters. At Moberst, we believe that
• 33:14
  creative thinking, advanced technology, and data are the
• 33:17
  core drivers of success at scale.
• 33:19
  If you're serious about scaling your business with precision and
• 33:22
  purpose, it's time to talk to Moberst.
• 33:25
  Visit moberst.com and let's build something category
• 33:28
  defining together. Hey founders, let me ask you
• 33:35
  this. Are you building with AI or
• 33:38
  getting left behind by it? If you're serious about scaling
• 33:42
  smarter, faster and bolder, you need to check out AI for
• 33:46
  founders.co. It's your go to resource for
• 33:50
  cutting through the noise and actually using AI to grow your
• 33:53
  startup. Each week, AI for Founders drops
• 33:57
  a sharp, no fluff newsletter, dives deep with expert guests on
• 34:01
  the podcast, and hosts monthly workshops where you can get your
• 34:05
  hands dirty with the real tools and strategies.
• 34:08
  Whether you're pre seed or series B and beyond, launching
• 34:12
  something new or just staying sharp, AI for Founders gives you
• 34:15
  the edge you need to win in 2025 and beyond.
• 34:19
  Hit AI for founders.co to subscribe to the newsletter,
• 34:24
  tune into the podcast, and grab your spot in the next workshop,
• 34:27
  and it's all free. Build smarter, move faster,
• 34:31
  think bolder with AI for Founders.
• 34:33
  Now back to the show, a couple of questions here.
• 34:36
  So this, this audience loves ideas.
• 34:38
  If you were to build a product that could quickly do $1,000,000
• 34:43
  in revenue, what would that product be?
• 34:47
  So I think if I were to start something in this space, what
• 34:52
  I'm really excited about and it's kind of similar to what you
• 34:55
  were saying, but for UI. So the whole idea of generative
• 34:59
  UI. So if like, let's say I'm going
• 35:04
  to an app of some kind, what if it just built the UI in real
• 35:10
  time for me based on what I needed, right?
• 35:14
  So I'm, I'm coming, let's say I have like an accounting app,
• 35:18
  right? Think of the complexity of all
• 35:22
  these accounting apps like QuickBooks or whatever.
• 35:25
  They're insane. And most of the stuff like I
• 35:27
  don't know, I don't understand, But if it could learn a little
• 35:30
  bit about me and where I was with my company and where I was
• 35:34
  with whatever the financial stuff that I needed and what I
• 35:39
  understood, it could build an interface that would ask me
• 35:42
  questions in the right way to get the data.
• 35:45
  And so it would have to tell me a little more than if an
• 35:47
  accountant logged on and they would get their own interface.
• 35:51
  So I think a platform that would essentially build user
• 35:55
  interfaces on the fly, I think that would be a killer app.
• 35:59
  I think, I think that's the future.
• 36:01
  Generative UI is just you build the app for what you need
• 36:05
  honestly. I'd love that man, that might
• 36:07
  exist. I was there's a great vibe code
• 36:10
  community in Denver, a great bi weekly meet up and I think
• 36:16
  somebody was talking about this where it like adapt you like
• 36:19
  your preferences. So I could say like, I don't
• 36:21
  want I don't want to ever see vertical menus.
• 36:24
  It's just it's annoying. It's too much on the screen.
• 36:27
  I don't ever want to see red buttons.
• 36:28
  I don't want to see red on the screen.
• 36:30
  Everything I want to be turquoise buttons and it start
• 36:33
  it will actually change as you go.
• 36:36
  I'm not exactly sure. Let me see generative AIUI
• 36:39
  designer because I heard that and I was like.
• 36:42
  There are some companies, I have seen a few.
• 36:45
  None of them have blown my mind yet in terms of what they do
• 36:49
  because I think the key is going to be less in colors and
• 36:52
  placement and more in like how it asks you the questions and,
• 36:55
  and how it asks you, you know, whatever it is that you need to
• 36:58
  do in that app based on your sort of understanding of
• 37:02
  whatever it is you're doing. Yeah, amazing.
• 37:05
  OK, If you were going to college, you're entering college
• 37:09
  as a freshman right now. Based on what you know now, what
• 37:13
  would you study? That's a good question.
• 37:21
  I definitely would not study coding.
• 37:24
  I would not study that. I would not go into computer
• 37:26
  science. No offense to anyone who did.
• 37:29
  But, you know, I think there's going to be, there will always
• 37:35
  be a need for coders because there's always going to be
• 37:37
  people who need to build the machines, but there's going to
• 37:40
  be fewer and fewer. You know, I think getting an
• 37:44
  understanding of, honestly, I think the more diverse and well,
• 37:52
  sort of rounded people are, the better off they're going to be
• 37:56
  in the future. Like people who are, you know,
• 37:59
  this is one of my big arguments because there's a lot of
• 38:01
  backlash against university education.
• 38:04
  And my general belief is you're not learning necessarily a skill
• 38:09
  always. I mean, I was a liberal arts
• 38:11
  major, but what you are learning is you're learning how to learn.
• 38:15
  You're learning how to think critically, you're learning how
• 38:19
  to think creatively. So I would go into something
• 38:22
  that just like grew the creative part of my mind, You know, I
• 38:26
  would study music or I would study writing, creative writing,
• 38:31
  or I would study art probably. And then something that was also
• 38:37
  like history or something that that you know, helps you to, to
• 38:41
  do that kind of stuff, but just grow your mind and have a solid
• 38:44
  understanding because it's going to be the, the future is going
• 38:48
  to be for, you know, a lot of didacts and people that are a
• 38:53
  lot more like polymaths, you know, then then it is going to
• 38:57
  be for specialists. That's my personal belief.
• 39:02
  If you want a safe route, I think Healthcare is always going
• 39:05
  to be safe. You know, there's always going
• 39:07
  to be people always going to be bleeding, you know, to put it
• 39:12
  bluntly. Thank God.
• 39:14
  Yeah. Oh OK, amazing.
• 39:16
  Drew. This has been so awesome man.
• 39:18
  Thanks for coming on the on the podcast.
• 39:20
  Where can people find you online?
• 39:23
  Yeah, I have a website, it's still in progress, but it's Drew
• 39:26
  falkman.com and that'll link you to anywhere you can.
• 39:30
  I do a lot of courses on LinkedIn, so by all means please
• 39:33
  connect there. I'm on some other social medias
• 39:35
  and stuff too, so look around. I usually just go by my name.
• 39:40
  But yeah, thanks for having me. It's been, it's been a really
• 39:42
  great, great time chatting.