Subscribe
Share
Share
Embed
Explore the evolving world of design with Cameron Craig and Keith as they tackle the challenges of complex, monolithic products and the critical role of human-centered design. Each episode dives into topics like organizational change, the future of design in tech, and the emerging influence of agents on user experience. Perfect for designers, strategists, and leaders, this podcast offers insights on adaptability, communication, and the strategic thinking needed to thrive in a rapidly changing landscape.
Keith (00:01)
Cameron! Hey man, how you doing?
Cameron Craig (00:01)
Hey Keith.
Long time no see, man. How you feeling?
Keith (00:07)
Getting better. I have this weird not COVID slash felt like COVID but not like it's been going around New York. So I might cough a little here and there but all good. We'll make it happen. dude, your lighting looks great.
Cameron Craig (00:19)
Yeah. All good.
Thanks. Yeah. It's, yeah. Yeah. I've been working with that a lot, trying to perfect my setup as I was complaining to you before we started recording, but camera, not great.
Keith (00:23)
It's like very, it's like natural daylight-ish, yeah.
Yeah, so sometimes the metering can
get messed up where it meters equally. It's average versus on like the center of the spot. It takes a while and then if you don't have like a very regimented lighting that's consistent, like if it's daylight and it's changing, it can mess with ⁓ all the auto settings and stuff.
Cameron Craig (00:52)
Yeah, I may have to reach over and adjust the light, the darker it gets. Cause you know, windows that way. Anyway. Yeah. ⁓ I'm okay. It's been crazy town, both, you know, work and, and family life. The, plotted some charts for us and you know, you got some crazy, I got some crazy going on with my family.
Keith (00:58)
Yeah, it's good. Yeah, how you doing?
How is things good? You stable? You like better?
Cameron Craig (01:22)
Yeah, yeah. mean, again, as you and I were just talking, like
Well, one, as we transition into thinking about business problems and solving business problems with systems thinking, I more and more become convinced that the knowledge, thoughts, and advice that we're talking about and the design space around how that space is changing and where the world is going in terms of the technology, the tool set, and ultimately the skill sets, we're like spot on.
So, you know, we're continuing to build tooling to change the role of design at the company that I'm at. But as a side project, a friend of mine and I took on some design work and some consultative design work to help another friend. And we made the commitment that we were not going to do any traditional design. We were not going to use any current design tooling. We were not going to do
Any type of standard practice where we sit down, draw a flow diagram, hang the interfaces off the flow diagram, go and review with the customer. were going to just take input in from the customer and work in a new, ⁓ AI based coding and interface development program. And we were going to build and test and iterate.
Keith (02:40)
Whoa.
Cameron Craig (02:57)
in real time.
Keith (02:59)
So you're basically like vibe coding the whole thing from start to finish.
Cameron Craig (03:03)
We're like vibe designing and it's building out all of the front end interface with all of the states, the state maintenance and everything. And as you're describing features and functionality, the tooling will actually figure out the next steps of things. It'll stub things out, right? Like you're not going to get an end to end workflow as an example. So imagine, ⁓ a kind of lame use case.
Keith (03:06)
I love it. ⁓
It's like next concrete
step.
Cameron Craig (03:33)
Yeah, you're developing a messaging platform inside your application. So think dashboard with the most recent messages, ⁓ exploded view where you've got every message. You've got an ability of sending group messages or a single message. And when you say, Hey, I need to add group messaging. Literally the prompt is I need to add group messaging to this system. And it goes and basically.
Keith (03:44)
Yeah.
Cameron Craig (04:03)
in the page that you're working on inserts the icon for group messaging falls the same pattern that it already did for single messaging, but add adds in the, the modal so that you can select multiple contacts and then just loads it. was like, we just with one prompt literally did like four hours of work in
Keith (04:17)
Cool.
Dude, four
hours, it's like months of traditional UX with like requirements and flows and like approvals. Holy.
Cameron Craig (04:35)
Six minutes, Keith, six minutes, including
the time that it took to render it into the current system.
Keith (04:42)
So what are you guys using Claude code or what are you what's your like stack of how you're kind of building all this stuff or is this like proprietary?
Cameron Craig (04:50)
No, this is fully off the shelf. ⁓ I don't necessarily want to pitch a platform because the other, the other problem that I have is I am, I'm playing with most of the current state of the art, if you will, whether that, and I'll just mention some of them for people to try out, like whether it's V zero or build.io or bill, ⁓ bolt, ⁓ or for sell. I I'm playing with all of them. I, I'm not going to plug one over the other for a bunch of different reasons, but, ⁓
Keith (05:03)
Yeah.
Yeah.
Cameron Craig (05:19)
You know, just imagine that you're in one of those tools and it's happening there. And then from that, you have the ability of exporting your bare code, right? Like your HTML, your JavaScript, et cetera, whatever made up the front end. You have the ability of dropping those things into a repo like GitHub or any of these other things. And then having your development partners who are doing the back end wire up, come in and basically work with the code that you've developed.
Keith (05:45)
Okay.
Cameron Craig (05:49)
and
Keith (05:50)
Cool,
so you have a human doing the backend important security. Make sure it's not, yeah. Yeah, okay.
Cameron Craig (05:55)
Any of those things. mean,
there's templating for some of it, right? Like you can go in and basically be like, I need to, you know, have storage. So like that everything isn't, you know, essentially dynamic. And when you end session, you either have to store things locally on in cookies or whatever, or, know, within your browser, ⁓ you know, like you said, I need social login. Like they have a bunch of these pre-made plugins for different things.
Keith (06:05)
Sure.
Cameron Craig (06:25)
But eventually this is not gonna be something that somebody could go from idea to production today. bet you in six months from now, bet you these systems will be robust enough to at least get you a performant and secure, very basic application that you could put out and not be worrying about. You're probably not like dropping that.
Keith (06:34)
You don't want to coat on pride?
Secure enough, yeah.
Dude, that's awesome.
Cameron Craig (06:54)
for your customer like Macy's or Safeway or a bank necessarily. I mean, it's honestly, Keith, even if I were working in a bank, I would probably author all of my stuff, export all of this into some repository that I could then pull from at work or yeah, or at the very least I would just cut and paste all my code and you know.
Keith (07:10)
Yeah why not.
Just sandbox it, exactly.
Cameron Craig (07:25)
my own repository into some development environment and
You know, I'm not handing off wireframes. I can tell you that right now.
Keith (07:35)
Dude, ⁓
so it's like being your own agency at this point.
Cameron Craig (07:44)
Hell yeah, dude.
Keith (07:46)
Yeah, think the more I see, because I was thinking about like, okay, the tools and tricks and kind of like what we were thinking about with like, formation conditions and like, running like a counter-insurgency to like improve the thing, suspend your ego, all that stuff. The bigger corporations like, they're not going to be, there's going to be a handful that are massive. Like I almost feel like Anthropic and Microsoft, like they're all going to kind of like merge, like they're going to own pieces of each other and just kind of be this like, ultra
monolith that's a know, pieces of each other in a way. But if the tools are getting better, at some point, it's going to get easy enough for people even to have like imposter syndrome to be like, you know, it's good, I'm just going to try it. then and then they're to go from dog or try it to like, Whoa, it can actually make this work where I can make, you know, a basic app web app in like a day versus like 60 hours or you know, so as the pace goes increases, it's like enables anybody.
Cameron Craig (08:45)
So suffice it to say this little science project that we're working on, ⁓ started, it's today, Thursday, we started Tuesday evening, about this time Tuesday evening. And we have the golden path for this rather complex piece of software that will facilitate ⁓ investor, financial investing between venture capital companies and hedge fund companies.
Keith (08:56)
cool.
Cameron Craig (09:15)
all the comms, all the doc storage, all of the deal making, the upfront deal. ⁓
Keith (09:21)
Yeah, due diligence
and compliance and
Cameron Craig (09:26)
They can plum those things in, but we basically have given them the place where they can create a deal room, load the docs, do all the comms securely, right? Like all of this will sit in a secure environment that they can control. Like I only want to calm with these people. I only want, you know, deals and the materials that are related to the deal in this particular repository. All of that.
not non work time hours over the last three days. like to your point, it's leveled the playing field and it's exactly what we were talking about in terms of designers are no longer at a place outside of the four walls where it's like, thanks a lot designer.
Keith (10:00)
That's crazy.
Cameron Craig (10:19)
Really appreciate the wireframes. The development team and I are gonna go off and do some things and we'll tell you when we need refinements to your design. Like.
I've never had a smoother customer meeting last night ever in my consultative life where I'm like, and they're like, what about this? Could you do this? Yeah, hang on a minute. And they're talking to us and saying, well, you know, actually we need, we need a, we need the ability to, to message certain groups as a mass, certain groups, like as we pick them and individuals, like I want to be able to go back and forth with.
Keith (10:38)
Wow.
doing it in real time.
Cameron Craig (10:57)
the other person on this particular deal and have it outside of the flow of, you know, the, the running thing, right? I want to have the ability of like, if Keith and I are talking about something important and it needs to be in the core, you know, calm flow, just basically boop, boop and have that go into like the running conversation. Yeah, no problem. And like, we're just querying and prompting inside of a set of
queries that we've developed and the things doing it, you know, it's so crazy. So.
Keith (11:30)
God. So were you
were you spinning this up on your own kind of discovering it? Are you working with somebody who's vibe code design stuff beforehand? You just kind of like.
Cameron Craig (11:39)
You got two
completely, ⁓ code ignorant designer business people, myself and you know, long-term, collaborator who, you know, used to work for me directly and then went to Macy's with us. Left Macy's went to Albertsons left Albertsons came to Amazon. Right. So like long-time collaborator and neither of us.
Keith (11:47)
Cool.
Cameron Craig (12:06)
can code where we're technical at the architecture level. Like we understand the, the infra that ends up being, you know, the underlying system, but we, we can't, I can't link storage to a database, to a visualization system.
Keith (12:10)
systems.
And why do you need to know how to do that now? It's like...
Cameron Craig (12:22)
I don't,
I mean, that's a problem I'm solving in my day job, Keith. So, you know, like, so we gave ourselves a challenge to basically see if we could figure out these tools, do something at rational scale for us, right? Which is consulting and getting to a place where we are not just defining the product. Static we're defining the product and all of its States and all of its front end and being in a place where we're going to go and meet.
Keith (12:47)
Yeah.
Cameron Craig (12:50)
with a actual developer who will do the infrastructure in the backend wire up either tomorrow or Saturday and literally see if, if, if our code is at the quality that he can take it in and actually start wiring up the functionality.
Keith (12:57)
Dude, that's nuts. Took a week.
guide. So how would you recommend somebody get started doing something like this?
Cameron Craig (13:13)
pay the 30 bucks a month for one of these systems and just go play. Like honestly, took us, it took us about three hours to get a environment up and going that met our needs in terms of the visual look, like the visual look and feel actually took some of the longest time because we're like, no, that's not quite right. and you have to kind of define the atomics of whatever it is that you're doing.
Keith (13:17)
Just kind of figure it out.
Yeah.
Cameron Craig (13:43)
And then once you do that, it, it just falls through, right? Like it knows it knows and keeps track of the, the, you know, call the action button is, is a totally stupid example. And that should always be this particular color, no matter where it is until you tell it to do some, all that, all that, all the typography, all of the like iconography. took us, like I said, about
Keith (13:43)
Thank you.
Yeah.
with a certain drop shadow or whatever. Yeah.
So the design design
takes the longest part in the upfront. And once you get the design system in place, you just like, OK, cool. Now we can start building the modules of what we're doing.
Cameron Craig (14:11)
Yep. Yep.
And this is again, like I want to make sure I'll take you on a tour at some point. ⁓ so this is again, where I think we need to stress systems thinking, design thinking and business thinking. Right. And so to do this well, you still have to think in systems. You have to.
Keith (14:20)
Dude, I gotta check this out.
Nice.
Okay, say more.
Cameron Craig (14:46)
Like I was saying, you have to think about the atomics first. Like, Hey, what am I trying to build? What is the style of it? You've got to define those things. And I, and for some things, I'm just throwing out the visual look and feel of the branded thing because these guys care about that. It's important that when this thing gets presented to partners and investors, that it looked professional. It you know, modern and, and like a solid financial institution, if you will. Right. Like, and
Keith (15:15)
You're asking for
money. It's got to look like it comes from money people like, yeah, brand.
Cameron Craig (15:19)
And the, the
idea creator is out of big four consulting and the collaborator, the person who is wanting us to get this to a certain point before they invest money is from a very, very large bank. Right. So we're not, we're not playing founder at this point. We're literally, we have like the two key players already involved because
Keith (15:35)
Okay.
But there's.
Cameron Craig (15:46)
We were brought in, you know, it's like, I'm going to get these guys that really understand design and product development to come in and, and take apart the problem space. And then basically build the software or build the design. First, they were just like, just lay out the flows and figure out, you know, the interface, and then we'll figure out how to get a builder. like, pause on that. And then that's when George, my long time collaborator is like, let's, let's not do this.
the way we would normally do it. Let's do it as AI driven and as dynamic as we can because for our day jobs, we're kind of like, got to figure this out, you know? And we're building something similar that is much, much greater in scale, right? It's not going to enable two dudes trying to figure out an investor website. It's going to be the thing that somebody like, say, Anthropic needs to build. ⁓
Keith (16:37)
Yeah.
Cameron Craig (16:44)
A custom knowledge base for their employees to train everybody as they come in on all of the previous science, if you will, for certain things, right? Like that type of an application is what we're trying to build or, have a system that enables, ⁓ somebody in a company like that to build a giant knowledge base as an example, right? They can build anything, but like that, that knowledge base at scale is not a simple problem to solve. So the tool that we're building.
Keith (17:05)
Right.
No.
Cameron Craig (17:12)
sits on top of obviously a very large cloud infrastructure. And we want to have people think about the application and then have the infrastructure come along with the interaction. Where the tooling that we're using right now doesn't necessarily bring the infrastructure along, it kind of stubs it out. And then, you know, we're going to need to find a human to connect all the infrastructure.
Keith (17:24)
Got it. ⁓
Well, it sounds like there's like a very specific business need in terms of like communication and like deal flow and like getting everything in place to talk about it, to have the discussion that's like legally or financially like copathetic and like auditable or whatever. So there's a very specific business problem that you're trying to solve that they've obviously had like a hard time trying to like get over doing it like the old way. So that I think is like the design and business thinking part where it's like, you're not just like
Cameron Craig (17:49)
You got it. You got it.
Keith (18:05)
Making some website, you're kind of like prompting, you kind of got this basic design system in place in three to five hours that was like, okay, here's kind of like the look and feel and kind of like the vibe for lack of a better term of it. And then you're like, okay, let's start with like, they're gonna have to have logins. You're gonna have to have like profiles and it's going to be like this kind of an app is what we're trying to do. So it kind of has context and you kind of go from there and then. Yeah.
Cameron Craig (18:31)
And
that's what I'm trying to get back to is like, to do this well, like, misnomer, I'm gonna hand anybody a tool and all of a sudden they're a designer. I'm gonna hand anybody a tool and all of a sudden they're a developer. I'm gonna hand somebody a tool and all of a sudden they're a founder, right? There's gonna be a percentage of the population that can do that.
But really again, where the systems thinker, and I don't just want to say designer because there are other, there are other roles. So I think I'm starting to genericize my language. The person that is a systems thinker. We'll go through and systematically start to define the world that they're playing in. Right. And for us, we immediately took apart. What is the problem space? Who is the customer that this is being built for and who is the audience is going through? And we were like brand.
Super important to find brand first, because like one of the things that you don't want to try and do, since this is a very visual process is reinsert that in after you've built a bunch of stuff. let's slow down. Let's take the time to define the visual system. All right. Next, like flow throughs, they're throwing out features. Hey, you know, people, people need to talk. People need, you know, group talk. They need to secure group talk, individual group talk.
Keith (19:50)
The old way.
Cameron Craig (19:58)
individual talk, like all these things. We're just like, okay, okay, got it. We're like, get to that. Let's stay high level, right? There's comms. There is docs. There are deals. There are contacts. And we're starting to just build these buckets of things, right? And then
It was funny because we go from the buckets. We're like, okay, let's talk about all the things that are in a deal bucket. As an example. All right. Let's talk about all the things that are in a contact bucket. All right. Let's talk about. Yeah. Yeah. Yeah. So from that, we literally got to like the information architecture, the navigation system, all that stuff within a.
Keith (20:44)
It just comes organically
based on the requirements of what you're explaining as your main buckets.
Cameron Craig (20:49)
Yeah. Like another five minutes, 10 minutes by the end, we're just like, okay, got it. You know? And they're like, you know, and then people are starting to be like, well, the nav system can't have all those things. And I'm like, the nav system is going to have almost nothing in it. It's literally going to be like contacts deals. It's like,
Keith (21:05)
Yeah, ducks,
yeah. All the basic bullshit,
Cameron Craig (21:09)
And then, then, you know,
we, took a dive into competitive systems in this particular space, right? This like investor enablement space. And they just overwhelm the user with so much information. Like you, you hover over a particular contact and you get this fold out splay of information on the contact. was like,
Why do I need that? And you know, of course the invest. Totally, it's exactly that, Keith. It's like this like you tap like you F10 and you get this exploded view of Keith, you get this exploded view of Cam and it's like, does anybody look at that stuff? And and so then, you know, of course, the the person at the bank is literally like, well, that becomes really important if I want to go through my ⁓
Keith (21:37)
It's like post Bloomberg terminal kind of thing. Yeah.
Yeah.
Cameron Craig (22:04)
all of my contacts and understand, you know, who, who knows about deal making in retail. And I'm like, well, wouldn't a smart querying engine, AK in this moment, AI basically just be like, Hey, John, it's Keith, it's cam, it's George, and it's Steve. Like, why do, why do I, why don't my making you the human being go process through all that? And you could just see the like,
And I was like, this isn't even really AI. This is a good search engine, right?
Keith (22:36)
hardwired.
It was AI to build the products to like, just facilitate the speed of development. I mean.
Cameron Craig (22:48)
So, so of course, again, you know, we're like, well, let's just try it, right? Like, let's tell it that we need a cli a querying engine that can break the contact into all these things. And, know, it brings up like a little like search box. brings up a filtering system. It like takes each of the. It's all been done before. So it takes like all the features that we, all of the header for each of the form fields we have, and it basically just turns it into a querying engine. Brilliant.
Keith (23:05)
It's all been done before.
That's amazing. mean, because in a way, it's like it's the best. It should pull the best of the best that people are used to right now. So think it just everything got so kind of commoditized from like an interaction and like a design standpoint, because it's just there's only so many ways you can design like all those things that that meme of like the start. It's like you start up one or start up two based on like if it's like the header and like the CTA and then like the layout is like two basic layouts that kind of like change a little bit because
the with the print looks the same because it's you know it's portrait and whatever it's a certain size so you're right I think at some point what ends up happening is like the designers who are really good aren't gonna want to work for agencies or big places they're gonna just be like I can do this myself and pay charge half of what these big places charge and make like 80 % margins doing it because it's so fast now that it's like why would you
state is a large corporation at this point.
Cameron Craig (24:17)
Well, I mean, and it's such low overhead too, right? it, it's all those things that you and I be like, ⁓ God, it's going to take weeks. Like we know what we need to do. Now we got to go and marshal the resources. who are we going to get to do this? Who's going to design, who's going to do the visual design? Who's going to do the interface design? we need somebody to sit with you and I, we need till to sit with you and I and like, do the information architecture. You don't need any of those people. Like it's.
Keith (24:20)
Yeah.
Yeah.
Cameron Craig (24:48)
It, it, it's working as fast as you can conceive the things, but the, again, where I keep coming back to the, the leg up for the systems thinker. I'm, I'm, I'm cause I'm in front of the keyboard for part of this meeting. I'm the scribe and I'm typing in, you know, what, what the person on the other end of the phone is telling me. I'm, I'm like,
I'll just create a query based on the things that they're telling me. They, they obviously know what they're talking about better. It just gets totally convoluted quick. And I was like, all right, I know what you're saying. I'm not going to type exactly what you're saying. I'm going to query the thing and I'm going to do it in the context that I understand. And I'm just adding a little bit of detail. And all of a sudden it's like putting the functionality in the right place. Like, geographically on the page.
Keith (25:32)
It's your own filter
and experience processing what they're saying they want, not just putting a brain dump in there to kind of get them back out what they really need.
Cameron Craig (25:42)
Yeah. And you know, that's what I'm saying. The people that think in systems, like start here, build on top of this. Now go to this. Edit away, you know, once you've established those things, but you really have to start thinking about this is on top of this is on top of this. These things are connected. And if you can't do that, trying to talk to a machine.
to get it to understand those things in context, it's just gonna be like explosion. Yeah. Yep.
Keith (26:12)
It's garbage in garbage out. Well, that
that's why you have people going crazy with AI and going having like psychotic breaks because they're feeding it nonsense. And it's just it only is as it only gives you back or reflects back what you put in. And it's it's cumulative. So it's the same kind of thing where if you don't know how to structure these ideas or kind of like see these things. I don't know.
You can teach systems thinking in terms of like their stocks and flows and you know, but I kind of wonder if how much, do people have it naturally based on how they see like some people who design who are just like look and feel really good at it, they're gonna like be really fast at the beginning part and then they may have to, it's like you have to kind of start with something simple and maybe it's about like what is the kind of core
resonant loop somebody is like innately vibing with that they can kind of learn to branch out of. Maybe that's the Trojan horse to kind of get them to systems think better the rest of it. You know what I mean?
Cameron Craig (27:17)
Yeah. Yeah. And I mean,
also, you know, expertise in a given area is still really important, right? Like if he's George and I, George is a really great world builder, right? Like he's a way, he's like a way better world builder than I am. And by that, mean, once he figured out all the branding, he then started to structure the main system that we're working in, you know, all of the different sort of redundant.
Keith (27:22)
Exactly.
guys, perfect for this stuff.
Cameron Craig (27:46)
pieces that are repeated, right? Like he built the basic pattern. And then as I'm now flushing out details in a given area, like I'm really good at communication flows. As an example, like I'm really good with like sender message receiver, like all day, every day, back and forth. And so I'm coming up with these tertiary features for this part of the system that, you know, I.
Keith (28:04)
Yep.
Cameron Craig (28:12)
They're pretty innate to me, but not to everybody else. And, you know, he's back working on kind of the deal flow. This is how something moves from point a to point B to C to D. Here's all of the like notifications that need to send when things move from one place to the other. And he's really thinking through like all the mechanics of that, you know, where I'm not as good at that. And, you know, it's
Keith (28:25)
The waypoints.
So it's.
So it's two
system thinkers who can go micro, macro, but each of you have either a macro focus or a micro focus that you do well. So it's like, that's the 80-20. That's cool too.
Cameron Craig (28:46)
Yeah. Yeah. So just,
yeah, it just goes to show that, you know, while we think we can drop a designer or a business person from one detailed industry into another, the more complex the, the interaction, if you will, that you're trying to designer, or in some cases, like with banks and compliance, you know, like the security, when we get down to the security stuff,
Keith (29:11)
god, yeah.
Cameron Craig (29:16)
We're going to need to bring it, like bring the person from the bank back in to talk about all of the security compliance before we start building that because.
Keith (29:22)
Yeah.
But if you have like
a skeleton and like a working body and like an armature around it. Like it's much easier because the AI is so complex now that you could just be like, okay, have a talk with Jose or whatever. And you have, you you filter it, put it to the AI. And it's like, you're not just putting it in there. It's like, help me build a spec to put the security into this kind of like framework that we're building right now. And it's like, okay. And then it's like part two.
How do I test this to make sure all these use cases are like bulletproof as much as it could be? know, test them yourself. How would you break it if you were a red team or a hacker? So it's kind of like this recursive, you know, thinking. mean, there's like some tricks you can learn to it, but I think having a pair person who you communicate really well with, I think is kind of like a secret unlock to all this too, because doing it on your own, it does like in the initial part, it sounds like there's a lot of, ⁓
like ramping up and kind of like domain knowledge to kind of like you have to kind of get to the initial like you said the first three hours is like a little bit of a slog to kind of get the brand and look at all the basic wiring kind of put together before you can start actually seeing a working thing and getting kind of the payoff to induce the flow. But yeah, I think it helps having somebody you can communicate well with.
Cameron Craig (30:46)
Yeah, I agree. I mean, I think we oftentimes underestimate the value of the human comms and the human connection between two people figuring out a problem. you know, while it may seem to a company or a customer to be more expensive, it actually is, you know, in the beginning, when you start at zero, it's obviously going to be more expensive, but having two people work on something
Keith (30:58)
Yeah.
Cameron Craig (31:16)
prevents one person from getting completely stuck on something. And at the same time, now we're diverging. Like he's on one path, I'm on another, and we're working in tandem at the same time. So you're going to get twice the amount of features developed tonight as an example, as you would if it was just me.
Keith (31:21)
Yeah, totally.
Yeah, we-
Yeah, and if you guys have a good communication rapport and you understand each other on a telepathic level, but it's easy to talk and you have an almost intuitive sense of what each other is thinking and what the vibe is, then that I think is important too. I think anything human-wise that you have an issue with, normal collaboration gets amplified with all this AI stuff. And if you do something well, it's just way easier and it goes way faster.
If it's like there's an issue, like you're going to find out real fast, that is an issue.
Cameron Craig (32:09)
Yeah. Yeah. Yeah. So it's a really interesting time, I think. And, you know, I've been processing a lot of this, like, honestly, it's a little bit like, what does it all mean? Right? ⁓
Keith (32:12)
Cool.
⁓ dude, yeah,
everyone's going, look at that right now.
Cameron Craig (32:28)
Yeah, I know. And I'm really empathic. I'm I'm I think I've become even since the last time you and I recorded it, I've become a lot more empathetic to that. ⁓ But it is like we've said before, this is not an all all hope is lost moment. This is a this is a iterate, learn, be curious.
Keith (32:48)
no.
Cameron Craig (32:55)
tap into the things that have made you successful in the past. And I now more than ever believe that systems thinking is actually the most important thing in this moment in terms of being relevant and in terms of solving problems. And you're going to have to layer on top like an innate curiosity and an experimental nature where you're just like, I'm going to try it. if it, if it doesn't like we literally were working on a deadline, but I'm like, we got to try this because the unlock
To our future time drop. If we don't have to figure out, and I don't have to go in source and I don't have to transfer all of this knowledge to somebody else, we're going to move really fast. And. You know, we went back and forth with all of the different tooling and then the process that we would go through. I'm like, well, if we did it all in Figma, like we're fast in Figma, we could probably design all of this stuff quickly.
Keith (33:28)
Yep.
Cameron Craig (33:52)
We could probably get to a place where we'd have a relevant prototype that we could show people, get them to a place of understanding and, and core and concrete understanding, right? Like we could still click through an actual model and show them things, but like we couldn't work in real time with the customer where the customer is like, well, what if we tried this or Hey, what do you guys think of this? We also would be in a cycle of ingest.
requirements go away and design review, take in changes, take in new requirements, go away and design. And I was like, I just think the closer we can get to code. The better. I mean, honestly, like my collaborator, George was really instrumental in being like, it may slow us down. And if it does, we can go back to Figma, but I think, I think we need to try one of these.
like AI enabled coding tools. Like what do we have to lose?
Keith (34:50)
Yeah, well, it's going to be slow at first.
Yeah, because you got to figure out your process and it's something new. So the innovation path is slow at first, but as you just keep going through more cycles of it, it just gets faster because you got more seed time and then you start getting more efficiencies because you're like, now I can do this.
Cameron Craig (35:08)
Yeah. Yeah. So that compression though, the one other thing that, know, I think you and I need to keep harping on is like these individualized roles that we all had, these lanes that we were in, it's really blurry right now. And, and I think, I think the people that think the biggest
Keith (35:22)
dude, they're gone.
Cameron Craig (35:32)
They're going to end up being the winners in this situation, right? Like you can teach yourself after a while, how to make better queries or how to start to break your things down contextually into something that an LLM can understand. that knowledge and that fluency comes quick, but you have to keep the bigger picture in mind while you're like, you, you said it really well. Like you have to keep the ability of going like macro, micro, macro, micro.
But having the entire system in mind as you're doing it, that becomes really important because it, the one thing I'm also learning is you can start to introduce deviation into your system. you're not careful in like when you're querying to like make functionality happen in this example. And I mean, it's the same thing in writing. I've seen it happen when you use an LLM to do writing.
Keith (36:05)
TechSculpture.
Cameron Craig (36:29)
You can start to insert deviation and the deviation gets wider and wider. If you're not eye on the entire system. And sometimes you have to take a break from the creation and go and do the inspection and be like, ⁓ wow. Like somewhere along the lines, this changed. And I've got to make sure that we don't cast us ourselves off into a different direction, like without knowing it, you know? So I think.
Keith (36:36)
You got it.
Recap.
Cameron Craig (36:56)
those become the new sort of tricks of the trade or the new like expertise that you need to basically continue to blur those lines.
Keith (37:05)
I'm wondering if there's a way to instruct how to convert systems thinking to leverage. Cause a lot of people who naturally kind of have that where they're like, they just kind of pull the pieces together from the hole. can go, it's like just all thinking like they kind of have macro micro or they don't. And then you can learn it. They can say what the kind of LLMs and like learning how to refine prompts and break ideas down into like more concrete structures. But I wonder.
Like, the flows and the site maps, I like doing the whole time just because it was helping me figure out the micro macro of what was going on. was giving me the conceptual shape of what was getting built and we're kind of like the buckets of things, you know? So I wonder if there's gonna be like a new wave of information architecture tools to just scale systems really fast. Because this kind of stuff now you're talking about, it's like, okay.
Cameron Craig (37:50)
Yeah.
Keith (38:01)
push a button, the environment's up the way I like it. It's like getting your tools in Figma set up where it's like, okay, I got like my little wireframe modules and whatever that I can kind of just pull stuff and I got like my break points or whatever. But even that, it should just be basically a standard now. It's like, I don't want to have to think about that. It's that's just like, it's tedious. It's like having to go like shave for 30 minutes. It's like, dude, I want this time back. Like, seriously, like I don't want to do this tonight.
Cameron Craig (38:27)
Yeah, it's a great point. mean, that I left that out, right? I was so excited, ⁓ you know, kind of to talk to you about the actual workings of this, this not just interface design, but this product design, you know, that we're doing, but you're right. Like I have two weeks of fid jam behind that. Right. ⁓ and that isn't time lost. isn't a waste of time. It was.
Keith (38:38)
Yeah.
Cameron Craig (38:57)
two weeks of human conversation with multiple people and the collection of the ideas and the translation of the ideas. Look, you're right. Like there goes, I'm now I'm blown out because the
Keith (39:09)
Did
your volume went down to that you had so I want to take an auto setting. How do your level sound when you look at you talk okay now it's getting better again.
Cameron Craig (39:14)
That's weird. Anyway. Yeah, they're all right. Yeah. It looks fine.
But, ⁓ that Fig Jam collection of, of both ideas and then ultimately pre-structure, right? Like, Hey, we need to think about these, like we set up our own buckets before we did a bunch of client and customer interactions and
Keith (39:22)
Fig Jam.
Cameron Craig (39:39)
We all contributed, right? Like our collaboration happened there and it was slow because it starts with a bunch of talking. Well, like, what do we know? Like, how do we know this? And, know, again, everybody's bringing their expertise in. You've got the person from the bank talking about finances. You get the person from big four consulting talking about, you know, sales and marketing and how you attract an audience, how you, you know, ⁓
Keith (39:49)
Yeah.
Cameron Craig (40:07)
What's the word I'm looking for? Like how collaboration works when it comes to deal making. And then I'm translating how collaboration works.
Keith (40:15)
Who's the customer
and how is it going to work across all the domains? Who's the real?
Cameron Craig (40:18)
Yeah. And I'm thinking
about digital systems at that point. Like, all right, well, how do we turn this into something that we can move into an interface and ultimately have, ⁓ some kind of a system to manage and maintain this stuff. Yeah. So.
Keith (40:32)
test.
But even that, it's like, to me, it's the most fun part, like the discovery part is the most fun because it's like you're learning all this new stuff. People who are normally in these boring silos feel like they're being creative. I made a quick joke, kind of half joke post about like everybody feels like Steve Jobs and they have Post-it notes because you're like, my god, I'm getting my ideas out and I can like strut move things around. And it's kind of a hack. you know, it's
It's equally as important because if you set that up properly, like the structure, it's important to get that kind of domain, those domains figured out because then it's just, the build gets much faster, much easier because you have like a lot of the hard lifting done because it's like, that's what builds your brand. It's like, is the vision? What is the value system? What like the real shit? Not like we're equitable and we're safe and whatever those things are. It's just like a lot of that is a corporate BS rather than, no, like we're...
We're secure. We're legit. Like this is high net worth. Like these are bankers. Like these guys, they have sat phones and you know, they have two of them because they got backups and they don't have time for downtime. Like, and they've got money to afford it. Dude, it's like the new Gordon Gekko's, you know what saying?
Cameron Craig (41:39)
Money never sleeps Keith. The money never sleeps.
Is your wake up call Keith Conway? ⁓
Keith (41:51)
Oh my God. But but yeah,
dude, I want to talk to you more about this later because I've had ideas for apps that I've wanted to kind of like do and I just I haven't had enough time to you have to put some seat time in if you like to kind of just like exit just play an experiment and just BS with it and just have fun. But anybody who's employed right now and who's like, this job sucks, but I don't want to leave because of recession or whatever. Like experiment, spend the 30 bucks on you
Claude and like for sale or whatever like there's tons of videos on YouTube on people who've done this. know and just try it and see what happens and keep it really simple like think of a simple idea like I would love to have ⁓ a contextual search engine that could just look through all my photos all my text all my stuff because I take a lot of screenshots to kind of like hard code the day into things and like things you see on Twitter it changes they delete it whatever.
Cameron Craig (42:47)
Yeah.
Keith (42:47)
So I've
been doing this for ages and like the search doesn't work that well on Apple yet. Like you can do it in your phone unless you, don't know. I want to have control of my own thing. want to run, I want to roll my own local LLM with like a smaller quantized version of DeepSeek or something. And then just be able to go through it fast. If I made this, where is that? Cause if you're editing video, it's like, I know I have this clip of this thing. Can you find this for me somewhere in the last 10 years that I was doing shooting and it's like, oh yeah, it's right here. Boom. There you go. And then it's like making you get faster with the things you want to do.
Cameron Craig (43:18)
Yeah, I mean, I think the other thing like the state of business right now.
Keith (43:25)
Dude, it's weird.
Cameron Craig (43:25)
I don't know if you can't have a
con if you can't have a conversation about this and you're not knowledgeable about the basics of how these things work, right? Like forget trying to necessarily apply this to your day job. Your experimentation at this point is just say you're knowledgeable about what it is and how it works. Like I have a hard time believing that these conversations are not happening in every corporate environment right now. And if you're the lone resource at this point, who's like,
I don't know. mean, that that's gotta be it. Like a detriment. Like I would imagine, like, I hate to say this and I, mean, honestly, I hope this isn't happening, but I, I also believe it probably is happening. If you're the one person who's like, Nope, you probably are going to be looked at as the like Luddite, right? It's you know, it's like, flat earther, like, hate to tell you this, but you may be.
Keith (44:16)
You're the first to go. I mean...
Cameron Craig (44:22)
You may be packing your stuff into the banker's box and seeing your way out.
Keith (44:28)
Yeah, well, it's hard. No, I, but I also think there's going to be restructuring that's happening and they're going to say it's because of AI efficiencies and they just need to cut, they just need to cut weight because the costs are too high right now.
Cameron Craig (44:30)
I don't want to do that, but I just, you know.
⁓ dude. mean,
ask me how I know this. Like.
Keith (44:52)
This isn't this is everywhere because you're still see like, dude, the jobs numbers have been fucking faked for the past like four years they did revisions it's mark to market data. And then this
Cameron Craig (44:58)
Yeah, I know.
I mean, that,
yeah, that's for real. That's for real.
Keith (45:05)
So
you want to talk macro micro. mean, what I was excited to talk to you about partially, this is cool. I dig that you're getting into this because I've been not there, but I'm like slowly right behind is like, dude, Intel basically getting a 10 % ownership from the US government now where they converted the grant from the Biden administration into stock. And it's like they initially went all in.
from you know only the paranoid survive to like hey we bet the government or someone who is going to be our next best cause we're going to build fabs everywhere now the government has a modular fab for like a war of bits not bombs and what's happening right now dude people are doxing people like people on x are docking people on blue sky because of the rhetoric against certain people who were murdered or assassinated or shot and whatever you know politics aside it's like
a holy war in like a MMORPG that's getting rolled together. And it's like, I see this and I'm like, my God, this is the cover of like, the hell is coming? This is why people don't see what's happening. And it's like, can't, you know, no.
Cameron Craig (46:10)
Yeah, it's not great. mean, for sure.
Like, and I found myself even, you know, as these things are unfolding, it's like...
And I had to check myself. going to just say it. My first inclination was, well, the left has finally gotten kinetic.
Keith (46:23)
We all do dude, it's human.
It
was a step change. Once there's like political assassinations, it's like it's like suicide bombers from Iraq and the GWAT, the global war on terror.
Cameron Craig (46:39)
And, and it's,
and like, we're always like, we're always talking about it in the news. It's like homegrown terrorism, right? And I don't want to live in that country. I really don't. But you know, there, there was that sick, very human side of me, the, you know, politics aside, anybody that feels like they've been wronged and their side finally grows some teeth and fights back. And you're just kind of like, well,
Keith (46:53)
Yeah.
It's a lizard brain
dude.
Cameron Craig (47:08)
Yeah, it's totally the lizard brain. was like, he had it coming, you know? And like that, that's a sick, sick place to be. It's not a good place to be.
Keith (47:10)
That's what it...
It's the amygdala hijack they teach you in social engineering, but broadcast on a mass scale through algorithms. it's it's hard, man. I can feel it in the city. People are getting like more ornery. They're not, they're paying less attention. They're just like acting like there's nothing around them in their physical space. I'm like, it's a, feel like Neo dodging bullets in the matrix, just try not to get like run over by some guy on his phone, stopping halfway on the stairwell to look down. I'm like,
Cameron Craig (47:26)
Yeah.
Keith (47:46)
dude, like, get the fuck out of the way. Like, what the hell? Come on. And it's just, I don't know, man. So in a way, everybody's human. We're all figuring this out. Like no one else the hell's really going on. Astrologically, these things haven't happened in like 6500. There's so many big, transits that are happening right now that it's like an unreal time to be alive. But I think that's also, there's an illusion of it's doom and gloom, but things are just
There are some tragedy, but it's like bittersweet. You have to find the way and you have to kind of keep the light on and keep the curiosity going and just try new stuff. Like the old way is not going to work at all anymore. ⁓
Cameron Craig (48:28)
Yeah. I mean, I got to say
like, I've been struggling to outside of this and a few other things that I do to keep my brain working and sharp, ⁓ turning wrenches or, know, like participating in different like think tank type things. I've been feeling a little bit lost, but I got to say like, and you know, why, like today was a great day because, you know, we had a bunch of success with this particular thing that we're working on.
Keith (48:40)
turner wrenches.
Cameron Craig (48:58)
But I haven't really been jacked about work and work life and all the things we were just talking about being curious and figuring things out. Like I haven't wanted to figure things out that are work related in a really long time. And this week has all, has been all about unbridled curiosity and experimentation in a very like, not low risk, but like,
Keith (49:04)
Nah.
Yeah.
Fresh air.
Cameron Craig (49:26)
pretty low risk environment where like we can't really fail. And even if we could, who cares? You know, like the learning is the most important thing right now.
Keith (49:31)
Yeah, doesn't harm. like...
Yeah, we're right now we're in between eclipses, right? And it's like Pisces Virgo. Pisces is the end of the zodiac and Aries begins. And that's where the Saturn Neptune comes together in February. That's a 6,500 year cycle. So it's like a brand new embodiment, but it's like Saturn is structured. Neptune can be like deception or delusion, but Aries is like war. So this is like a structured holy war people are being pushed into. But again,
Cameron Craig (49:51)
Mm.
Keith (50:06)
I still think about it only takes like a handful, a small percentage of people to like start the change that other people seem like, oh, it's like all of a sudden it's easy to like the crowd takes over kind of thing. Or it's like the, you know. So I think a lot of people are also disenfranchised with, I send out hundred applications. I use AI to help me. The recruiters use AI to filter the things. Like nothing, it's like giant.
Cameron Craig (50:17)
Yeah. Yeah, for sure.
Keith (50:33)
It's like everyone's throwing every color possible and it's all just brown. And it's like, this is terrible. This isn't helping anybody. And it just makes people more ⁓ risk averse.
Cameron Craig (50:41)
Yeah.
Yeah, I mean, the synthetic
nature of things or the potential synthetic nature of things doesn't feel great. But again, it's like there is some synthesis of our knowledge base, our greater human repeatability, if you will, and taking advantage of that to make your life simpler, easier or better.
Keith (50:51)
Ahem.
Yeah.
Cameron Craig (51:09)
That's one way of
looking at it. You know, like you can get really caught up in, everything is synthetic and I'm participating in the synthetic nature of everything. you can frame it and basically be like, look, I'm going to take control of this situation and use these technologies and this synthetic stuff to enrich my life and give myself some time back to do things that I care about. Right. Like
Keith (51:34)
that you want to do. Yeah. Family,
kids, relationships, health, the real value based food groups.
Cameron Craig (51:41)
And I miss picking up tools at times and on the back end having a stack of wireframes. Sure. Like I loved those moments where you and I would spread things out all over the floor and be like, look at this. You know, like, move that over there. I miss that. But at the same time, like what I realized I really have been missing is the transition of idea in my head to tangible deliverable. it's like,
It's been such a slog to get that over the last 10 years that like to come full circle, it's like that leveling of the playing field and not being reliant on somebody else who controls my ultimate destiny is ultra freeing.
Keith (52:24)
think that's what's breaking a lot of people's brains is that they're so used to the system relying on it a certain way, like 401Ks and PTO, whatever. it's like, that's not really going to last a while. It's all in transformation right now. if people didn't want to have any accountability, then it gets even harder because you're trying to rely on somebody else to tell you or show you how to do it.
Cameron Craig (52:27)
I think it is too.
Yeah.
Yeah, and maybe not to go like full lizard brain or full evolution brain, but.
The humans that survive anytime, anything transitions are the ones that learn to use the tools, right? Like the people that had fire, they outlast last, the people that didn't quite figure out fire, the people that like turned stone tools into iron tools ended up dominating the people who had the stone tools, right? Like, I mean, it's, it is, it is, and
Keith (53:12)
Yeah.
You
Adaptability is key. It's right under the Macy's deck.
Cameron Craig (53:30)
You know, being in a place where you've got the idea and now you have the tooling to make the idea actually work, or at least, you know, free your brain.
Keith (53:39)
be open.
Yeah, it's at work where I'm at now. It's been helping people see how not to be like, like what happened with the computer? Why did it break? Like, why did my Wi-Fi turn off? like, I don't know, maybe hit some hockey. What happened? I'm like, it turned off. If this thing happens again, just look to the Wi-Fi, turn it back on. I'm like, it doesn't matter. Like, it's not this isn't like your credentials are getting stolen or this isn't like a security thing. It's like some
bullshit Microsoft Windows update that you can't turn off that might have done this because like it flips some bit in the bio. She's like, what? I'm like, just trust me. I know I'm like, I'm watching our about our watching our six like I built a standing desk and as I was locking up that one night, I saw there was the door panel was the I'm like, this is basically broken into because I never noticed it. I'm usually you know me I'm like always like hyper like, yeah, all the time. Yeah, I'm like
Cameron Craig (54:17)
Just go with it. Yeah.
Yeah, you're like looking. Yeah, yeah, you're vigilant.
Keith (54:36)
It turned out it was always kind of messed up. And I'm like, how did I never not see this? But anyway, so, you know, I'm always looking at that stuff because it's important. Like the red teaming part is something I realized this past couple weeks too, is like not turning my brain off from not fitting in predefined groups. You're not a hacker, you're not a red teamer, you can't pen, but it's all bullshit. Like, fuck it. Like, whatever. Like, I'm always red teaming stuff with my mind. I'm always like breaking it in my, not to like be nefarious, but to like,
This could be stronger and better this way, because you would find new business cases or you find new like design cases or how to like make the whole thing work better. It's like, so Vegas was Vegas. was like, Vegas was pissing me off because I'm like, they can make so much money if it is improve the customer experience rather than trying to like milk the bottom, like 30 % of like the economy from people. And now they're wondering why they're not making any money. And it's like, well, here you go.
Cameron Craig (55:27)
Yeah. Yeah, I who
gets to fly to London in a few weeks to talk about making certain types of gaming more
Keith (55:41)
Dude, that's gonna be amazing. Are you kidding?
Cameron Craig (55:43)
⁓
more adaptable.
Keith (55:47)
Can't fire up that LLM and make a demo, do you? Versace? Okay, so how do we make this gaming more adaptable?
Cameron Craig (55:52)
God, Keith, it's like, there's part of it that I'm like, I can't like, this feels a little bit like those moments when we were at Macy's and was like, how do we increase the bag size from X to Y? How do we get these people to try a new service and increase the bag from X to Y? Like.
Keith (55:55)
Do it.
So
that to me is, it's important to have a measurable goal because you don't want to like to spin your wheels and play around and not have anything to show for it. But it's also detrimental to be so like parts per million are so measurable on everything that it's like you miss the point of what the unlock really is. Because it's like fine going from 80 to $100 sure it's like whatever 20 % whatever.
Cameron Craig (56:40)
Yeah.
Keith (56:44)
30 % increase. But it's like when I was at Estee Lauder, they realized that if they had virtual beauty advisors in like the lab coat and they had all the knowledge, like the cart doubled like more because it was women talking to women. It's like someone had an affair. The dog just died and they're you know, it's like you got to buy this and this. They're fucking buying all kinds of crazy shit. And it was just like they were like, what the hell? And then do this. It was on flash. Right. There's three guys who built this company.
And I told one of the head marketing people, like, you're not going to do this. I'm going to tell you to do it, but I'm telling you because you're not going to do it. She's like, really? What? I'm like, you guys need to go buy this company right now. You need to go turn it to HTML5 and then you need to white label it and give it to every other freaking like Clarion's, Clinique, give it to everybody and then take, you know, like a subscription fee or like a do something. Cause then if it makes it better for everybody and it's proven to work, which it has, I'm like, you just built a freaking ATM machine. And she was just like, what? I'm like, yeah.
I'm like, this would cost you guys like 10, 15, $20 million right now. Those guys would be psyched. Three dudes out of like MIT, they'd be like, I made money. And for you guys, it's like the artwork in the hallways of fucking Estee Lauder costs more than like what the acquisition costs would have been. But they couldn't do it because the brand team hated the production team. It's just like, I'm like, I'm like.
Cameron Craig (57:48)
Yeah.
Yeah. Yeah. Well, I mean, you lived
through that with me at Macy's with the stupid beauty box. Remember that? And it was like, like the brief comes down. We need to modernize our, our beauty brand. And we're like,
Keith (58:08)
my god, yeah! It's like...
We can't
find the thing because the search doesn't work.
Cameron Craig (58:22)
Well, and then, and then remember on the back end, dude, it was, it's exactly what you described. The merchandising people were like, we can't go back to SD latter and tell them that they need to throw this in the box. I'm like, why not? And they're just like, they won't do that. It was like, and even if they do, then we've got to figure out like, and then it became, I mean, they were thrown around words like fed Phil. And I was like, you guys are worrying about like checking this stuff into an inventory management system. Like,
We're going to run an experiment. We're going to run it for like six months and see what we can't do that. We can't tell Estee Lauder that a channel is going to go away. And I'm like, guys, call them and tell them what you're trying to do and see what happens. And of course they call Estee Lauder and Estee is like, yeah, we're in. Like how much, how much of this nonsense do you need and where do we send it? And Dylan's like, you send it to 680 Folsom street in San Francisco. And like, I'm, I'm not joking, Keith.
Keith (59:06)
Of course.
Cameron Craig (59:20)
months of like boxes showing up from different like beauty company clinic, you know, and it's like,
Keith (59:25)
Dude.
Cameron Craig (59:29)
he, I mean, he, he was a consultant and he literally got fired for that because it didn't, the boxes came, they didn't get checked into some inventory management system. And somebody was worrying about like all of these things that you don't need to worry about at the point where you're trying to do experimentation. You are not worried about like loss prevention at that moment. Like if half that shit like disappears, it doesn't matter. And they're just like, we can't
Keith (59:30)
Cause they knew.
my god.
Dude.
Yeah
They were pissed they couldn't take
credit for the idea. That's why they threw them under the bus. Let's call a spade a spade. Let's call politics politics. But these are the kind of things. Maybe what would be awesome is if people... So there's like ⁓ a feature on the website that we have, which it's still like beta, whatever. You can record ⁓ audio and send questions in. It'd be amazing to get people's like political...
Cameron Craig (59:59)
Yes, Keith, that's exactly what it was.
Keith (1:00:23)
bullshit they have to deal with and send them in and then have us a red team it how would you fix it how would you address it how would you not get fired and shot a dude I would love that we'd be so good at this too because it's like it's like
Cameron Craig (1:00:28)
I would love that.
That could be like a segment every week if people actually like submitted stuff. It'd be so fun to do. So fun to do.
Keith (1:00:40)
I think I think we should
make video clips and put this out because Dylan Red team the thing he's like, no, you send it here. You don't send it to New York, you send it to San Francisco, we're gonna deal with it.
Cameron Craig (1:00:51)
Well, he kept showing up
to my office and he's like, dude, like, and, know, can't believe I'm admitting this, but, uh, it'll be, it'll be good marketing for this idea of like people send us our ideas. So he would come into my office every day, largely for the, is it going to be okay? And I was like, I don't know, but we should do it anyway. And the best thing ever was we were told.
Keith (1:01:14)
Figure it out.
Cameron Craig (1:01:19)
by our boss, our mutual boss. I'm gonna be surprised if a thousand people sign up for this. And I was like, like to the red teaming thing, I'm just like, challenge accepted by friend. You know, so, well, it was with a bunch of conditions, right? And.
Keith (1:01:31)
Dude, this... Yeah, we'll get 10k of these fucking things. Because it's free shit!
Yeah.
Freaking for purchase. Dude, do you know how crazy, Let me tell you, finish your thought. tell you a story about freaking for purchase after this.
Cameron Craig (1:01:44)
We...
Which I'm sure is probably the same story. So we, we put this thing out. We, we met a, they were not called, ⁓ influencers at the time because recognize this is now nine, 10 years ago. and we seeded the idea with a couple of bloggers. We had a hundred signups in an hour.
Keith (1:02:07)
We talked about that too a decade ago.
Crazy.
Cameron Craig (1:02:19)
We had 2000 signups in two hours. We had 19,000 signups in about six and a half hours. And then we were told to shut it down.
Keith (1:02:32)
That's right, I forgot about this.
Cameron Craig (1:02:33)
And we're like, shut it down. I don't know. Hang on. Let me go call down to the IT department and see like who we need to file the form to shut the form off. And, and, and Dylan had called the guy and he's like, yeah, I can do it. Hang on. You know? And he's like, well, hold on a minute, just hang on. And then we both like set the phone down and we're like, we're still watching the thing go up. And he's like, I'm to shut it off at like 25,000. So we call the guy back and we're sort of like, all right, shut it down. Well, he ended up.
Keith (1:02:41)
Whoopsie.
Cameron Craig (1:03:03)
Going home for the day. So we had to track him down. So it went the entire night and into the morning. By the time we came back, it was less than 24 hours. had 75,000 people signed up.
Keith (1:03:08)
my God, perfect.
Dude, why would you turn that? It's like you have demand. It's such a strong signal. And this is kind of like.
Cameron Craig (1:03:19)
And we even said,
not everybody's going to, we had a database now of, of 75,000 interested people that we did not have before. And like the first thing that we did was cross check them against current customers. And like half of them were current customers, half of them, you know, half of the half were probably very active. like not a lot of upside to the company to really market heavily to these people. They're already active customers, but that's still left like almost 30,000.
Keith (1:03:39)
Yeah.
Cameron Craig (1:03:47)
not particularly active or brand new who, who we followed up with some of them were like, why have you never shopped with us before? was like, you guys seem like you're so far behind Sephora. Like, why would I bother? Right? It's like the first interesting, exciting thing they've seen come from this like dying company in like more than 20 years, you know, shut it down, Keith, shut it down.
Keith (1:04:03)
Basically.
Yeah. mean, the core premise, God,
it's clear when the turf war stuff, the problem was, we did coming from the group that we worked in, there were too many of those examples that came out of that. It's like, every couple months, it was like some new fresh hell somebody else had to deal with. It wasn't a bad thing. was like, they were just pissed. They didn't think of it first. They couldn't figure out how to like
put it together with like minimal resources and that that's. That's what we're really good at and what we were trying to get the higher ups at Macy's to see was that if you have this huge catalog that was actually accessible and searchable, which it wasn't because it was just old infrastructure. Then you had this huge playground and even the influencers. It's like why don't you turn the least model on its head where influencers? Could like least part of the capacity.
And they have a whole bridal thing and it's like bridal gets them into like the baby and then the shower and then it's like the kids first prom or whatever. It's like there's a whole longer customer lifecycle that you can build based on families or whatever. it it becomes very clear. But it's the leadership that doesn't think in systems either. And it's like if the leadership doesn't see or understand, just like, I'm making my EPS.
Cameron Craig (1:05:24)
Nobody thinks in systems. No one thinks in systems.
Keith (1:05:37)
I'm getting my bit, so fuck it, I don't care. And it's like, that's why, you know, these things die out eventually.
Cameron Craig (1:05:38)
Yeah.
Well, I mean, I think you've hit on something else, is, you have to have a calibrated risk profile that's like looking at the long game, right? It's like one of reasons why, you know...
done well, you end up with something like Bell Labs, right? It's like a monopoly. They're not going away anywhere, but they're thinking about problems that are 10, 20, 30 years out. And they're, they're generally right, you know, on those timelines. But when you're trying to be right on a quarter by quarter basis, you're just rolling the dice. I mean, you're literally rolling the dice. You're like, I think this one's going to be good.
Keith (1:06:03)
Yeah.
it's not about being right. Yeah, it's
I mean, it's like, ⁓ Brian nickel, ⁓ Starbucks right now. It's like, changed like the Starbucks by me. Like it's cheap enough. It's a large coffee relative to like the hipster stuff. That's like $9 now for it's like, you can't even get a large iced coffee. It's cold Bruce. They can upsell you. And I'm like, dude, this is, I'm like, it's annoying. I'm like, look, this guy, because he got pinched and hit for like the climate BS because he was flying a jet like three days a week from like California to like,
Seattle or something. mean, this dude needs to do interviews of people on the plane. And it's like, that would be huge. It's like the guy, what's his name? Sal or Sal, like in the jet business, he has a YouTube channel where he sells 5, 10, 15, $30 million jets to like the Prince of Saudi Arabia or whatever. And he breaks it through the business. And he's like, you know, this is a fixed cost. It's your hard cost. You want 20 people, but you know, whatever. And it's it's kind of entertaining because you get a peek into a place you wouldn't normally see. And I'm like, dude, that guy.
Cameron Craig (1:07:11)
Mm-hmm.
Keith (1:07:21)
I started to tap into this, you know, and you obviously like, if people are going to say BS and like going to tirade, it's like you screen people but you there's such a those kind of things people aren't doing and they could be doing to like reinvigorate like, there's actually like real blood in this old stodgy company now that it didn't used to be. But I don't know, man.
Cameron Craig (1:07:42)
Yeah, yeah, for sure. I mean,
Maybe, ⁓ one thing to end on from my week and back to the red teeny thing. And by the way, like we should encourage people to go find that form and record something. Cause that would be a brilliant segment, but I used, ⁓ I.
I can only hear and then, sorry, go, go on the ridiculous journey through my mind. ⁓
Keith (1:08:14)
It's all good.
Cameron Craig (1:08:16)
I only, every time I think about the last big op that we did, right, the shaming op, see what all of your innovation has done to the customer at checkout. Remember that one? And all I can think about is, know, ⁓
Keith (1:08:24)
The bookmark?
yeah, that was the best, one of the best ops.
Cameron Craig (1:08:39)
the end of die hard. Cause you used, know, well, the weather outside is frightful. As, as you know, you're doing the zoom in, like it's like the parking lot, the entry then like we're at the check, you know, the checkout or whatever. I used that video on Tuesday, part of it. I didn't show them whole thing, but I used part of that video on Tuesday to talk about that op because we're doing a very similar thing where I work.
Keith (1:08:45)
a kick in the head.
Cameron Craig (1:09:09)
currently where let's show the execs how this actually works. And, you know, we've had two experiences, we've juxtaposed them. One, we had a principal engineer actually use our stuff.
Keith (1:09:26)
I from Macy's or your stuff at... Okay.
Cameron Craig (1:09:28)
Currently, So
imagine somebody has to set up cloud infrastructure.
We put the person who we thought was going to be an absolute dead ringer in terms of like, yep, you go here, then you do this, and then this happens and you connect these things and then this happens. then, you know, like, look, you've got a database. Look, now you can do data visualization, right? Subject matter expert, principal engineer.
Keith (1:09:53)
Subject matter expert. Satya Nadella
doing his AS100 demo on fucking Microsoft Excel like 30 years ago, whatever. Yeah.
Cameron Craig (1:10:01)
Yeah, so imagine that.
⁓ it ended up being multiple sessions over multiple days, elapsed running time to get this very basic setup, which involved taking five or six services and stringing them together to get to an end result. was more than nine or 10 hours of elapsed time.
Keith (1:10:25)
God.
Cameron Craig (1:10:29)
Plus she couldn't complete the transaction.
Keith (1:10:31)
my God.
Was it because like a bug or because she couldn't do it or what happened?
Cameron Craig (1:10:36)
Literally couldn't do it.
Keith (1:10:39)
Was it just because the system wouldn't enable it? so it was like a, I actually couldn't do it.
Cameron Craig (1:10:44)
No one knows. I mean, so imagine now somebody outside of the company trying to do the same thing. So nine or 10 hours is the expert who works for the company trying to do it.
Keith (1:10:46)
my god.
Dude, this is like the most monolith thing ever, where it's like, yeah, you take 10 minutes, no problem, 10 minutes.
Cameron Craig (1:11:01)
So then we did it again this week, different problem space, one product set up to get you to a place where you, it's an app builder, right? The thing that I'm doing now, George, it's an app builder.
My designer figured out how to do it, but, but, but we couldn't actually get her demo up and running because for some reason, after she completed the task, the system decided that whatever she had set up was insecure and shut it all down and shut her out of the system. So she could no longer use the system.
Keith (1:11:41)
It backhacked her. It thought she was hacking it, basically.
Cameron Craig (1:11:43)
Yes.
So then we're like, all right, well, we'll just grab, you know, the principal, ⁓ product manager who also knows the space pretty well. And we'll, we'll do it in real time. Like, you know, my designer's like, you know, it'll take like an hour. We can get it all done. There's like nine of us on the phone, all of us with knowledge around, you know, some are all of the system, but the product manager really knows the systems. Couldn't do it.
Keith (1:12:07)
So it's just too complicated, it just wouldn't happen.
Cameron Craig (1:12:10)
Just like, I mean, and watching her do it was amazing. She had probably 10 tabs open. It's like, and then I got to go over here and copy this. Hang on. Let me do that. Okay. Okay. Wait, now I got to move my credentials from here to here. Okay. Let me go do that. ⁓ now I got to go over here and tell this system that it actually needs to grant this other system access. I'm looking at this. like, there's no way I'm like, and what does this enable us to do? And the product manager's like,
That's storage. Like that's what it took to spin up just being able to store the result out of this thing. She's like, yeah, like. ⁓
Keith (1:12:50)
What were you trying to get it to do? you trying to get it to do something complicated?
Cameron Craig (1:12:54)
We're trying to get to a place where we can actually have a working environment to build an application and then use this tool to build the application.
Keith (1:13:02)
So this is just like the tool set up in the environment to build the thing that you actually want it to build.
Cameron Craig (1:13:08)
right.
Keith (1:13:11)
Dude, like, okay.
Cameron Craig (1:13:14)
So, so,
so two things, one showed the video of, Hey, here's what happens when innovation goes wrong. Or here's what happens when you turn a bunch of product teams loose to do their own thing and they work in silos and they're doing the right thing by the customer for them, but they are not cognizant of all of the other things that we're doing right by the customer in the same like workflow. And my boss is like,
Keith (1:13:16)
You
Cameron Craig (1:13:35)
This is brilliant. We need to show like we need to record the next one of these things and show it to like the execs. I'm like, yeah, you want people with their heads down on the desk crying? Like this is the way to do it. He's like, I want to do it. I'm like, great. Let's do it.
Keith (1:13:46)
Dude,
having video skills, a little bit of video skills and storytelling, it's like, you're so dangerous because you're just like, can't deny it.
Cameron Craig (1:13:55)
Well, we're all we're
flying to Seattle with a bunch of, you know, Insta 360s and we're to do the same thing. We're to have like the one overhead. We're going to have like the one to the side. We're going to have the one on our screen, you know, like, yeah.
Keith (1:14:06)
Dude, it's so easy
to jam sync this shit and like premiere a final cut. It it all those things that now like it's it wasn't. It was in its infancy when we did we had to do it by hand but for now it's just like you just put it together.
Cameron Craig (1:14:18)
I mean, it's
the three 60. can just say start at this point, all cameras and. And then I can just be like, show this. Okay. Show this, show this, you know, it's brilliant. Super quick.
Keith (1:14:29)
And can
and then you can go back and there's tools now that let you extract text out of the speech to create like a script. So you could just copy paste that drop it in L.M. or whatever and be like find me these pieces. Find me this make me a new cut. Boom boom boom. And it's like it's so much faster now. Yeah.
Cameron Craig (1:14:44)
Yeah. It's pretty amazing.
All right. Take us out, Keith. What do you got? It's on the backend. You're usually good at summarizing all of our high points.
Keith (1:14:50)
Man well leash
Let me show you hang on screen. ⁓ I have to share this. Hang on. Can you see this?
Cameron Craig (1:15:05)
Yeah, I can see it.
Keith (1:15:07)
Hang on. So if you want to leave us a message, I haven't tested this, but it will work. If you go to themonolith.tv slash contact, you can send us an email or you can record a message. So record us a message. Tell us what company dilemma, political bullshit, whatever issue you're having. You don't have to say the name of the company, people's names. Give us context. Give us like what you guys are doing. Just tell us what's up.
Cameron Craig (1:15:21)
Nice.
Keith (1:15:36)
And then we'll play it and then we'll tell you how to fix it. And we'll give you this would be great. This is going to be fun because I think this is something that we're really good at this. like it's we want to help you help yourself in the process. But yeah, the monolith.tv slash contact, you can record a message. Yeah, that's what I wanted to get into. But what was the other thing I was thinking about? The last bit, the Banksy. Did you see the Banksy that got like
Cameron Craig (1:15:50)
Yeah.
Nice.
Keith (1:16:06)
They spray painted over. There was a barrister walking in front of it. Let see if can find this real quick.
Cameron Craig (1:16:08)
Now.
⁓
I did see that, I did see that.
Keith (1:16:14)
It looked like the second slide in the deck that we did.
Cameron Craig (1:16:18)
Yep. Yep.
Keith (1:16:26)
presentation. I can't wait to screen.
I'm in the wrong browser. Yeah, one of the first things we were showing in our deck, I gotta find it and pull it up, but it was basically all these barristers or like judges in the UK and they all have these like old wigs on it. It's like we've always been doing it the same way and it's like it's so classic. And I'm like, I saw this.
Cameron Craig (1:16:54)
yeah,
I've used that like the whole, just because we've always done it this way doesn't mean that we should continue to do it this way. Like everybody in the same wigs. Yeah. Yeah.
Keith (1:17:03)
Yeah. Yeah.
I'll have to find that for later. But I was thinking about that. But yeah, dude. This is the time. Like
There's a business case being there's something about being a hacker being a business person being a systems person and Finding out how to pull it all together. I think I don't know what it is yet, but there's something here and I think the implementation we had before were like The corporation part. I almost think it's it'd be more fun to like help people with their problems That's I really hope people like submit audio But I think the real next wave is gonna be helping individuals spin up smaller teams to compete against the bigger companies
Like you just said, you couldn't even get a freaking environment set up. And it's like, if it takes that much work for someone who's like a pro and works in the company, who's been there a decade, it's like, at some point, the AI is just going to figure out how to do it faster. the innovation will move somewhere else.
Cameron Craig (1:18:04)
Yeah. I
mean, and again, I think it will figure those things out, but we are going to need to tell it what, and again, I know if we, if we ran an AI script that be like, cam says systems thinking too much, but at the very bottom of the systems thinking chart is defining new mental models. And like, this is the moment where all of the existing mental models are going to get tossed. Like,
Keith (1:18:33)
%
Cameron Craig (1:18:34)
That is my top level 50,000 foot learning over the last six months. If you are not rethinking the mental models, you might as well just stop because none of this stuff that we have deemed super relevant is going to remain relevant. And in fact, a lot of the things that we've done that we thought were incredibly relevant and we sort of built our immutable laws around in terms of the internet, everything's in beta right now.
Keith (1:19:04)
Yeah, dude, 100 % and like all this crazy holy war shit going to like digital ID to like, it's all cut. It's all the foundations were laid the last since from like 2018 till now, we're going to start seeing what the new wet next wave looks like next February. And then, you know, now it's like, one of the best examples I heard was, okay, this is going back to the past to get all the tools, the best of the best of the tools that you had to bring those forward.
It's not the point anymore. But it's like, okay, you got to make your run. Start making the run now. And it's like, just fuel it with curiosity. Try new stuff, especially if you're employed now. Just like just try it. You know, especially you know, our kids in college, like I know it's like, Oh, what am gonna do for a job? It's like, dude, you're in school, like you have time to mess around. And you don't have to worry about like health insurance and bills yet. Like leverage the time to get ahead. Don't get caught up in the social media bullshit and everything.
Cameron Craig (1:19:58)
Yeah.
Keith (1:20:03)
unplug from that, put the time back into this. you're going to see, I really think you're going to start seeing a lot of like really big, small companies. Big meaning like large revenue, high margin, small person companies because it's like once you kind of figure it out, you get that like the right customer profile, the right product market fit with the right like business case. It's like, all right, cool. You just kind of like keep feeding it. Just grow it,
Cameron Craig (1:20:26)
Yeah,
yeah, I'm gonna go figure out some 3D printing.
Keith (1:20:32)
That's the next one. That's why I think Intel is important to, I mean, dude, there's a lot of stuff going on in the background. I've been thinking about that. I didn't want to veer this conversation away from systems thinking too much, but it's a brave new world and there's a lot of opportunity. there's still a lot of, there's room for everyone to grow. It doesn't have to be zero sum in winner take all. there's, trust me, there's ways to leverage the change. So yeah, man.
Cameron Craig (1:20:57)
Yeah. Stay curious.
All right, Keith, as always, absolute pleasure. Yeah.
Keith (1:21:02)
Yeah, Kim, dude, always a pleasure. Yeah,
I'll have to try and get a new one in sooner than later. I know it's been kind of a wild couple weeks for everybody. Yeah, likewise. Thanks, man. It's part of the course, It's like, know, zombie apocalypse training. What are you going to do in New York? It's like, seriously feel like it's like a GTA level living in this city.
Cameron Craig (1:21:09)
Yeah, I'm just glad you're feeling better, really. Like, you know. Yeah.
Yeah. Yeah. I
hear you. I mean, on our end, it's like things move slower out here, but it's like, my son lost his mind and his appendix in the last like six weeks. So, you know, we're been dealing with that fun stuff. I have to talk about the, yeah, he's good. I'll have to talk about the medical establishment at some points. It's another fascinatingly weird space to be in.
Keith (1:21:33)
Yeah. How's he doing? Okay.
dude.
Coming yeah, okay, that'll be like Maybe it would be like how do we red team and read business case? Different verticals like it'd be a good segment to unlike. What have we seen?
Cameron Craig (1:21:56)
Dude, there's
so much ripe for possibility there. Yeah, write that down. That's good thought.
Keith (1:22:05)
Red team insists think different. All right, Cam, always a pleasure, All right, Later.
Cameron Craig (1:22:08)
All right, I'm gonna call it.
Thanks, Keith. All right. See ya.