KP Unpacked

Your Four-Year-Old Will Never Learn to Type

• KP Reddy

Share episode

Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.

0:00 | 46:37

What if the keyboard is already obsolete and we just haven't told anyone yet?

In this episode of KP Unpacked, KP Reddy and Nick unpack eight takeaways from KP's breakfast with who he calls the smartest person he knows in AEC. The conversation spans how kids born today will interact with computers the same way we interact with abacuses, why LLMs finally let humans skip the translation layer entirely, and why construction is already sitting on the most valuable physical AI dataset in the world without realizing it.

The deeper thread is about empathy. AI is galvanizing silos, not breaking them down. Everyone's getting more efficient in their own lane, but the 30% waste stays because no one is asking how their decisions affect the carpenter on the other end. KP floats a new term to replace "master builder": the project success manager. Not someone who tracks facts and reports delays, but someone who leads the project so every party wins. Then they pivot to what it means to invest in this era: stop looking for the best tech, start looking for people who can lead markets. There are only three voices leading AI right now. Why?

Key questions answered:

  • Why will KP's four-year-old view a keyboard the same way we view an abacus?
  • What's the difference between transcribing voice and actually understanding it?
  • Why does AEC already have the most valuable physical AI dataset and not know it?
  • Why are LLMs making silos stronger instead of breaking them down?
  • Has an architect ever asked AI how a design decision affects the carpenter? (The answer is no.)
  • Why is CRUD software mostly busywork and what replaces it?
  • What's the difference between humans in the loop and experts in the loop?
  • What is a project success manager and why does it matter more than a project manager?
  • Should KP hire someone with zero construction experience but amazing leadership ability?
  • Why are only three people leading the AI market when there's so much talent?
  • What does Brett Taylor have that most technical founders don't?
  • Why does KP predict Lovable gets acquired?

If you're in AEC wondering why your tools are getting better but your projects aren't, a founder trying to figure out what market leadership actually looks like, or an investor trying to update your mental model for what attributes matter most right now, this episode will challenge how you think about empathy, silos, and what it means to actually lead.

Listen now.

Join our AEC Summit - https://aecsummit.co/ - for a full-day event featuring developers, innovators, operators & business leaders and the massive new tech transforming the built environment!

Hair, Ratings, And Going Viral

SPEAKER_02

Hey Nack, how's it going? KP, there's a rumor out there.

SPEAKER_00

They cheated on you.

SPEAKER_02

There's a rumor going around that your ratings dropped off precipitously last episode. Is it true?

SPEAKER_00

It's directly correlated to my co-host's hair.

SPEAKER_02

Well, good news this week. We're back in action.

SPEAKER_00

Like, why'd you why'd you shave your beard off? I was digging the beard.

SPEAKER_02

Oh, really?

SPEAKER_00

Yeah.

SPEAKER_02

That was a recommendation from my wife. Okay. It was a little scraggly, you know. Yeah. I don't grow great beards.

SPEAKER_01

Yeah.

SPEAKER_02

But my hair I'll keep.

SPEAKER_00

It's funny the stuff that people like ping me on. Sometimes it's about the actual controversial content, so to speak. I don't think we're that controversial, by the way. I think just people are like, I don't know, they look for things to be controversial about, right? But it's funny how I pose I posted your text to me in our Slack at zero, and everybody just thought it was so funny. Like that one that, like, oh wow, Nick actually listens to the pod when he's not on it. That was the first reaction. Oh, like he actually listens when he's not on it. That's funny.

SPEAKER_02

Gotta keep people on their toes. You like auditing the pod. For the for the for the audience, my text message to KP was something along the lines of the previous guest who shall be named Andrew Ackerman, does not have hair that can compete with the regular host, co-host on this podcast. It was something to something to that effect, but it was somewhat derogatory. Yeah. So we we won't read exactly what said.

SPEAKER_00

And then I posted in all zero slack.

SPEAKER_02

Love it. Love it. Did did Andrew have a particular response to that, or did he happen to miss that one?

SPEAKER_00

No, I think it was just like, yeah, I mean, I don't disagree. I I saw him in New York this week when I was up there. And he was like, I didn't realize that was a prerequisite for being on your pod. He was like, I would have maybe not joined if I thought that that was like that that was a qualification.

SPEAKER_02

Got in, got in, gotta know your show, gotta know your audience. Our audience shows up for the hair.

SPEAKER_00

Yeah, and I think I I think our uh continual conversation about the hair is just trying to drive more YouTube viewership. Because if you're listening to this in audio, you're like, what is he talking about? Like we can decorate a hair. I gotta go check it out.

SPEAKER_02

We gotta get our YouTube numbers up, YouTube numbers up. So yeah, I'm glad I'm also glad there's a reference to Berman Wallace in there. Cause I mean, that was, I mean, I agree. He's the he's the king of great hair, and it's good to good to give a shout out to someone who sets the pace for everybody else. Yeah, my idol.

SPEAKER_00

I I do I do like what he's doing lately. He's kind of doing these well-produced videos, kind of like outside on like a sofa kind of or a chair. But it's you know, we're just kind of like doing Zoom, you know, as a word virtual. He has like he's built, I mean, in my perfect world, right? Like having like an aesthetic of how he's set up, I think he's doing any and it's edited well, right? It's got, I think he's got like a two-camera setup, and he's got someone like actually producing it. And I really appreciate that. I mean, there's there's some days where I think about our stuff, like in the world, if you're gonna do something, do it right. 100%. Yeah, yeah.

SPEAKER_02

Yeah, I actually think um the content's good too.

SPEAKER_00

The content's good too.

SPEAKER_02

Yeah, well, I'm glad we've we've transitioned to something serious. Uh I think the my my one comment on his videos that I think he's being super smart about is I think the future of communication in the technology world is going to be going to be oration

Why Real Video Beats Generated Text

SPEAKER_02

and oration that you can't fake, right? So whether it's spontaneous live stream videos, right, that only a human can do. Or, you know, just a really authentic, you know, single, single shot video like like he's doing, where he's just you know, sharing a unique idea or perspective that where he has, yeah, he has a unique perspective on something because he's he's he's sitting on a perch that is viewed, you know, is is maybe a slightly different has a slightly different viewpoint than everyone else because of his previous exposure, his current exposure and through his investments, etc. But I I think he's like a little bit forward-looking on that because yeah, in a world where text is abundant, easy to you know, to generate large amounts of text content, and but it's hard to produce, like, yeah, real, real authentic, authentic video with with a human connection.

SPEAKER_00

It's it's really interesting too. Like, also where I give him credit when you were like, hey, I can't make it last week. I thought about doing a solo pod. And then I was just like, I don't know how to do that. It's tough, yeah. I don't know how to do that. People, I mean, I I think it takes, I mean, you know, he's in Southern California, he's in Venice Beach. Hopefully, he's getting some coaching and training by people. I doubt, I mean, I'm he's a very authentic, like charismatic and thoughtful guy, right? But I I think like if I had to go do if I had to do a solo podcast staring into a camera, I would have to go get some training. I don't think I could just do that right now. Like, I think that is a skill that is very, very hard to do well, to do well.

SPEAKER_02

I mean, it feels like acting.

SPEAKER_00

It is acting. I mean, it is, right?

SPEAKER_02

I mean you're you're perfect, you're in some ways, you're performing for an audience. But I I think the the art of it is where you know where you start to get good at it is it doesn't feel like you're talking to an audience. You're just taught, you know, you're you're uh you know, you're you're you're not performing you're not performing, but it takes a long time to break through where everything just feels like a performance. It feels awkward. Like you're getting no feedback, you're just talking into a camera, it just feels like everything is magnified and they're just massive performance. So yeah, I I'm with you. It's hard to, it's hard to do. There's a there's a few there's a few people on social media who've mastered that style of content. And I from what I can gather, it's if if it's like a I think the best way to think about it is if it's a video, if you could record yourself and create a video that you can send to a friend or group or a group chat, and that's the entire intent, is like I'm just gonna record a live reaction to send to my you know my family or my friends, really low stakes, but a live reaction to a specific event. It's I mean, that that's pretty much what you want to do, and it's the it used to it should have some form of entertainment value, is is is mostly what it should be.

SPEAKER_00

Is anyone in your family using this app called Marco Polo? No, so people in our family use it, and it's exactly that you're recording a video for your friends, right? Like, hey, here's what's going on with me today, blah, blah, blah, blah, blah. And it just goes out to your friends, and then they can like share their day, right? And so it's just like meant to be this authentic way to try to keep up with your friends, right? And sometimes it's like literally like, you know, hey, carpool was horrible today. Like, I like I need to talk to someone, but you know, people are in meetings and we're working and all that. But I think it's meant to be this in the moment, and I think that's hard. I don't think I can do that. Like, I don't think I could record a video to send to you of myself. Hey Nick, how's it going, man? Had a rough meeting.

SPEAKER_02

It's gonna, it's gonna take some practice.

SPEAKER_00

It's hard.

SPEAKER_02

I think that I think so. If I was if I was media training you on this topic, what I would do is I would prompt I would prompt you on something something humorous. Because when when you're when you're able to like starting with hypothetically starting with the topic that is about hair, right? Yeah, like good example, like you don't have to perform, you're just reacting, you're making a joke about you know, about me or about whoever. And for that reason, it short circuits your brain and you're not performing. But if you have to like give an immediate reaction for, you know, the drone ploy pro core acquisition and it's formal and it's you know, if if it feels like you have to be immediately flip a switch and be serious, then it's gonna seem super awkward and take you forever to get there. But it yeah, if if I was like, hey, record a video making fun of me and you know, my scraggly beard for for for two minutes, I think you could come up with a pretty good video.

SPEAKER_00

Yeah. Speaking of which, like I think I, you know, we've talked about this. We're both high introverts, right? I mean, we you've talked about it. I think you just have like, can you just have like a website about like your speech and some of the stuff you're doing.

SPEAKER_02

In a past life, I had a coaching

Introverts, Feedback, And Finding Rhythm

SPEAKER_02

business that I where I tried to help introverts with this particular topic, yeah.

SPEAKER_00

And it doesn't mean you can't do these things, it just means it's like the energy expenditure is much higher than others, right? And so I was in New York this week at a NASDAQ event hosted by a company called Lumber that does like basically payroll management for subcontractors for self-performers, and they had me do kind of a keynote about zero, about like the state of AI and all that stuff, all that good stuff. And I had two people in the audience, Devin, my son, was there, and Andrew Ackerman was there. And you know, their point was like, because there's an audience and you get a reaction, right? When I, you know, make a pro core jab or an autodesk jab, people laugh, and they're like, that's when you get going, right? Until that first kind of funny cracks the seal, so to speak, they're like, then you're just like you're just running, like you're on fire, right? No one would ever know you're an introvert. But until you get that first laugh, it's a little dry. So I have to get those first laughs in early, right? To be productive. But it was you know, it was really, really cool. This group, it was probably about a hundred or so subcontractors, GCs, mostly CEOs in the room. And it's the second year Lumber's doing it. I was a keynote last year and a keynote this year, and it's a really, really good event in the in the fact that like no one from Lumber is really on stage, it's really more their clients and people associated with the industry. And I think it's a great example of you know, make your customers rock stars. You don't need the stage, right? You're there to support. I think they do a really good job of that. But it was very interesting in hearing everybody's AI journeys and what they're thinking. And and the the the name of my talk, you know, part of my talk was this idea of you know, if you really want to do well in the world of AI and exceed as a firm, as a person, it's about empathy and understanding. Like success is about empathy and understanding, especially in construction. And I think it really resonated with the group. I got like a lot of really good questions about it.

SPEAKER_02

Really neat. That's really neat. So I think one post that you wrote over the past week is going to take us through the rest of the pod. Okay. It's a lot of points to cover in this particular post. So let's just let's just jump into it. So the post is

Kids Won’t Type Like We Do

SPEAKER_02

I was at a breakfast today with a person that I believe to be the smartest person I know. I don't know how that could be possible because I wasn't in New York this week. He just so happens to be in AEC. Here are a few takeaways for me. And you go on to list eight, eight bullet points essentially of takeaways from that conversation. So, yeah, let's start, let's just run run through them all. I think they're all they're all quite interesting. So the first one is my four-year-old will view a keyboard the same way we view an abacus. Expand on that.

SPEAKER_00

Yeah, I mean, I think like the idea of the human interface, right? We we've stopped in a way, right? Like these AI chat bots, right? How much time are they spending on human interface design? It's not, it's a box, right? There's not a lot of interface design happening. And I think my understanding from some folks is that by the end of the year, when you're talking to Claude, OpenAI, and chat um, chat GPT, etc., it's not actually going to be transcribing your voice, it's gonna be understanding your voice, which is very different, right? It's not transcribing. So it might say something weird like, hey, you sound really stressed out, Nick. Like, you know, it'll be much more conversational. And and my friend, he brought up like it was kind of deep cut. I I forgot the which Star Trek movie there is, but there's a Star Trek movie where they go to a planet that is Earth, but back in time. And so they he goes up to computer a Mac, I think either Spock or one of them goes up back to I faintly remember, he was very quick to recall it, and he walks up to a Mac and there's a mouse, and like Spock picks up the mouse, it says computer, right? No concept of like a mouse, right, or an interface. So I think that's I mean, we're getting there so quickly, right? And I think once you start to realize that this it's gonna be about talking to the computer, not clicking on keyboards, not using a mouse. I I think it's super interesting. I mean, one of our engineers at Zero, he claims that most of his code he writes does via voice.

SPEAKER_02

Yeah, seems to be a common common theme.

SPEAKER_00

Yeah, yeah, which is wild. So yeah, I don't think I mean he's four. My my my son's four years old. I mean, I don't why would he learn how to type or have a computer?

SPEAKER_02

As just as an extension to this thought, it it it feels like potentially for the first time, I mean, at least in my working life, that in the in the next three to five years, we might not need to be working in front of a desk or like it like inside inside out of desk, yeah to do the to do the same type of work that we're doing today, which you know most people would classify as knowledge work. Just like you we won't need to sit there and type type into a you know into a keyboard. So this is it's an extension of the point you're making, but if it's already happening with with engineers, with with software developers, yeah, but maybe not, you know, it doesn't extend all the way to you know to to every single knowledge worker, I think it I think it will. I don't see why it wouldn't. And maybe that removes our our maybe removes the need to be in front of the computer from nine to five.

SPEAKER_00

Yeah. I I mean I I think 100%. I think I think what the the societal thing of that is I think we all need a third space, right? I think we all need spaces that are interesting to us. And maybe it's not like this boring aesthetic of an office, right? Like you're, you know, I've always wanted, you know, my in my dream house, which we also talked about this with my friend, like having a study with no desk, like just chairs and like lots of books, right? But like, where do I put the computer? Right? You'll you'll immediately like the minute you throw a monitor in that aesthetic, you screwed it up, right?

SPEAKER_01

Yep, yep.

SPEAKER_00

I just a couple Barcelona chairs, a nice table, a record player. And then, oh, well, where am I going to put my desk for my monitor? Like, no, like, I mean, so maybe we can get to better aesthetics.

SPEAKER_02

Yeah, very, yeah, very interesting. All right, point two, somewhat of an extension of this because it's about language. The power of LLMs is that we can know and speak to the computer without knowing its exact language or the language it's interpreting the the tech, the spoken text with. This is because it now speaks our language for the first time.

LLMs Let Humans Skip Translation

SPEAKER_02

We don't have to translate.

unknown

Right.

SPEAKER_02

Expand on that.

SPEAKER_00

So if you so if you think about what are LL, I think this is where people get really sideways about LLMs, have become, you know, kind of call it the peanut butter productivity. People think you can spread peanut butter on anything and it makes it taste better. And I think LLMs are becoming like the peanut butter productivity. And I and I think that's kind of missing some of the point. What we're what we're really saying is back in the day when I first started programming, I started in basic and then Fortran and then all that stuff, right? So we we've developed a series of languages that literally are just like layers of muck on top of the compute, right?

SPEAKER_01

Yeah.

SPEAKER_00

In order for the to to make it easier, right? So then we went from command, went from my mom started with punch cards, right? COBOL and punch cards and stuff like that. And then I was part of the command line, you know, beginnings of DOS. And then we added the mouse. We started adding like all these layers of interface for to make the to be able to connect with the compute as a human. We had to create all these things, and of course, you know, the screen and the mouse and Apple and everyone else. That's why we've had massive adoption to the compute. You know, there's a famous quote from the CEO of IBM where he says something like, you know, computers are amazing. One day we might have like 10 in the world, right? Because at that time it was like magnetic tape and it was like you had to have a special room and all this stuff, right? So we had to build all this interfaces to make it accessible to all people. And now what we're saying is like we don't need to do all that because now the computer knows how to talk to us. Right. And so that's I think the missing, the thing that people kind of miss when you're working with LLMs, you're you're really trying to train it on how you want to talk to it and how what you you know, you're giving it the knowledge so that when it talks back, it has a corpus of knowledge that is accurate and informative to you.

SPEAKER_02

Very interesting. Point three the AEC industry needs LLMs because we are a business of physics and physical. AI is very good at this. LLM is good at translating your words to do physics. Yeah, let's unpack that.

SPEAKER_00

Yeah, I mean,

Physical AI And The Real Deliverable

SPEAKER_00

I think the missing point, you know, there's all this like all these startups getting funded about physical AI, physical AI. And I found this company that if you're a like, let's say you're an architecture firm, you could actually sell this company all of your drawings and they will pay you for it and to feed their physical AI model. So we're like all these startups are getting funded billions of dollars for physical AI, right? We already have it as an industry, right? We already have it, right? Because we literally live in the world of physic, physical world, right? Then if you take like the physics of it all, we're in the physics business, right? How does a beam behave? How do you know elevations like it's all physics, right? It's physics and materials. So I think everybody's getting over-indexed towards the artifacts of us building stuff versus us building stuff, right? The drawings exist to build something, the specs exist to buy stuff, right? And I think a lot of the use of LLMs, like people are missing the point a little bit for our industry, that really it's going to be about physics and physical AI more than anything, because ultimately that's what we're doing. You know, it's it's like you the arguments I've always had, even during the days of BIM, which are now dead, have been that like drawing architects and engineers' deliverables are drawings. That's what they believe. Many of them believe this. And it's like, no, the deliverable is the building, that is what your deliverable is. But your drawings and specs and BIM, that is all an ends to a meaning. And then I'll tell you the most successful architects and engineers understand this. They understand, like, it's not about the plans, it's about the building. And I think when the design community gets really mad at like, why are GCs more trusted than we are with the owner sometimes? Like, I've seen owners pick the GC before they pick the design team. And it's like, well, they have great understanding and alignment of what the indeliverable is. The contractor knows the indeliverables of the building, the owner knows that.

SPEAKER_02

So you think it's rational to make that choice and choose the GC before the designer?

SPEAKER_00

I don't think it is. I don't think it is. I think people do because it's, you know, I mean, it depends on the architect and engineer. But I think we've created these silos, which by the way, not to go too far on a tangent, I think AI is actually galvanizing these silos, it's not expanding the silos. We're all getting much better at doing our work in our own little silo.

SPEAKER_02

100%.

SPEAKER_00

Yeah. We're we're the collaboration is going down because we think our new collaborators in LLM, it's not another human.

SPEAKER_02

Well, yeah, I mean, to and to to be fair, like one of the one of the theses we've talked about on this podcast, I think, really for the past two years has been AI is the return of the master builder. And the definition of master builder is it's single entity, one person. Right. Right? That's that's like the beauty and the promise of AI. So like this is the state of affairs, and uh it feels like we're yeah, we're continuing getting closer to that. I don't I don't see where that course corrects into a different direction, frankly.

SPEAKER_00

Yeah, I mean, think about it. Like, if we had an AI, whatever form it takes, right, on this podcast, and it was our third, it was our third host, right, was actually chiming in. That's very different than me being on a podcast with another AI, right? And I think what's happening is people are high, you know, and it look, it's like co-work. And where you co-pilot, we're using all these words. And I think the minute we say, hey, our my ideal partner to get stuff done is an LLM. What that does is that makes people double down on siloed thinking, siloed decisions, which is not our industry. Right. And I think part of my talk the other day about empathy and understanding is it's as much about understanding as it relates to how does my decision affect you? The big perfect example I made was oh, an RFI shows up, the architect gets the response back to the contractor at the last minute. Now a trade contractor has to stay late and miss his kids' ballet recital. Like that's not okay. Right. But but it's not like the architect's being malicious. It's just the architect doesn't know. Like all of it in a team dynamic, your decisions affect every party around you.

SPEAKER_01

Yeah.

SPEAKER_00

But I think in construction, we just we don't have this ability to that. And I think with AI, it's getting worse. You know, like I love it when people do this stuff, stuff with LLMs, like explain it to me like I'm a five-year-old. How about like talk to AI about your drawings, not about you as an architect, but how is this going to affect the carpenter's ability to build this? That that question has been asked by an architect zero times. Zero times. And I feel like there's this, there's this doubling and tripling down on, I can just work in my own little silo with an LLM. I don't need anyone else, which I obviously believe is the wrong idea.

SPEAKER_02

Do you do you think that I mean, is your is your theory because of the language breakthrough that LLMs present that someone will be able to decode and translate that effectively at some point in time, right? Like is that does that get better? It might get worse, you know, in the in the in the near term, but do you see that being a a language barrier that we can break through, or is it something else?

SPEAKER_00

Yeah, I think I think it's a lot of things. I just say I I think my point is like the I kind of view LLMs as the Retta stone for me to communicate to others. Right. I think I wrote a blog post the other day, a couple of weeks ago, about the five love languages and what are the like five love languages of construction, right? But think about it, you could go into Chat GPT. If you know your wife's love language, which I'm sure you do, right? You could just in LLM say, hey, I really need to break this news to my wife. This is her love language, rewrite it for me, right? Yeah, it will do a great job. Yeah, right. But that requires that you want to have empathy and understanding with your wife. So it's it's there, it exists, right? But I think you know, some of the stuff you have to, I mean, look, construction projects are just lots of negotiation, and you know, there's a weird power dynamic, right? Where if if I'm the GC and you're this I'm paying you, you do what I tell you to do, type of stuff. And I think I think it can be better than that.

SPEAKER_02

Okay, moving on. Other than design tools, most everything we do is CRUD software. It is not needed. Expand. We talk, we've talked about this a bit.

SPEAKER_00

Yeah, we've talked about it before. I just I I was talking to someone even just today, right? About how much

Why CRUD Tools Feel Like Busywork

SPEAKER_00

time they're spending updating Procore. And they're like, it's just like a governance thing. It's not helping me do my job. Like, I'm not waking up in the morning going, man, like I need to jump into Procore so I can figure out how to do my job better today. Yeah. And I'm about, I'm just it's not just Procore, it's anything else. Ask a salesperson about how Salesforce helps them do their job. What they'll tell you is like, well, no, I use Salesforce because my boss needs me to put stuff in the system so they know what I'm doing. Right. So I think same thing. I think with with everything we do in terms of project, you know, project management, construction management. We've just built, I mean, uh cred software, it's just a bunch of tables, right? You know, I actually asked this group the other day, I was like, hey, what do you think the top tools are that you use to use do your job? Excel, email, WhatsApp. Sounds about one person said pro core, not one person of it. So so why why is that?

SPEAKER_02

Because we're a fully autonomous these are the most basic Lindy forms of computer software and uses, right? Email, which is represent representing the internet, Excel, which is representing super basic computer software, basic forms of computation, right? And then WhatsApp, which is like the most basic primitive form of direct communication.

SPEAKER_00

Yep. So why is that? So let's talk about it. So a lot of people say, I can't deploy AI because my data is not clean enough, it's not structured enough, which is the exact opposite thinking, right? That's the whole benefit of AI, right? So I think we had to build all these CRUD tools because I'm not going to read your email, Nick. You need to put the information into the system so that I can have structured data to understand what's going on, right? Same thing with Excel, right? I need you to deliver structured data in the way I want it structured, not the way you want to get it structured. So my point being is now that we live in the world of AI, and clearly, this is a lot of the stuff we've built, right? Is I don't care. I don't care for the AI in an email. Upload an Excel spreadsheet. I mean, I got this debate about schedules and Gantt charts and Pert charts and last planner. I'm like, guys, it's just a bunch of tables. It's just a bunch of tables. Master format, uniformat, omni class, Kobe, all these things are just schemas and tables. And the problem is in the past, we had to somehow decipher how these fields it all connected. And we just don't have to anymore. So I think that's kind of the unlock is to what does all this crud stuff exist? It's like, oh, I need you to get all your stuff structured in a way so I can use it. Now it doesn't matter.

SPEAKER_02

Number six, this industry is about people. These aren't humans in the loop, these are experts in the loop. I like this one.

SPEAKER_00

Yeah, it's just, I mean, I think that's the I mean, uh, I think it's fairly self-explanatory, but I think this diminished

Experts In The Loop Over Humans

SPEAKER_00

idea, because we're a very old industry, we're a very large industry. I mean, there's some brilliant people out there, right? That will never consider, you know, they'll never say they're they're brilliant. I mean, I was, you know, like my friend I had breakfast with, right? We started talking about like blast harmonics and stuff. And I was like, you know, I was I was asking him, like, hey, what are you vibe coding? Dude, he's vibe coding blast harmonics. Are you blah? Are you vibe coding blast harmonics? I mean, exactly, right?

unknown

Exactly.

SPEAKER_00

And he's he's using AI to confirm his PhD thesis, because when he did it, the compute didn't exist to actually prove it.

SPEAKER_02

Well, that is like all these all these mathematical theories are being proven for the first time with these right.

SPEAKER_00

So he's trying to prove or disprove his his PhD.

SPEAKER_02

Great, great use case. Incredible.

SPEAKER_00

I mean, you you I'm more gonna say who it is, but you know who I'm talking about. It's like, of course, like, of course he is. Yeah, coding a CRM. Absolutely not, right?

SPEAKER_02

He's he's not asking Chat GPT for the love language of how to communicate to the contractor. He's he's underwriting blast blast harmonics. Love it. Great. Okay, you already kind of covered this one, but point seven was AI will not disrupt the industry, it will disrupt the people that lack empathy, understanding, and ambition. Yeah, you've already kind of pulled on some of those threads, so let's skip to the last one. So the last one is many AI tools will be very effective for companies, making them increasingly more efficient. It doesn't mean that we will shed the 30% waste, however, the silos will just be stronger than ever. And again, I think this is like

Stronger Silos And The Master Builder Return

SPEAKER_02

related to the master builder thread. But I mean, so so let's let's talk about. So the silos will be stronger than ever. I think there's justification that if you if if you actually go back to your comment on email, WhatsApp, and Excel, the problem with some of those tools, I mean, and maybe Excel in particular, is that it's an analog static document, right? And that was always some of the prom promise of software and SaaS was like one single source of truth that anyone can log into and get a real time, you know, real-time dashboard overview of their business or whatever they're working on, and be looking at the same thing. When you're looking at a static Excel document, if everyone's communicating, if everyone's interacting with Claude, getting a static Excel document back, and hypothetically, you know, someone makes a change to it, you don't see those changes. So we're back to we're back to the early days of you know, analog, you know, closed systems. And so I could see, I see where you're going. Like we're we're in this, we're potentially in this world where communication actually gets gets worse because the tools got better. And I'm like, let's just talk about like what, yeah, what are the paths to fix this other than someone's intent, someone's personal ambition, their empathy, their intent. How do we how do you like solve how do you solve this problem technologically and get and get everyone on the same page and maybe even push them, motivate their behavior to do the things that are going to actually push the industry forward?

SPEAKER_00

Yeah, I mean, it's so I think it's kind of like that interesting trope on the internet with VCs, you know, waiting to see the unicorn solo founder, right? And that it's interesting, right? I'm sure there is somebody that'll do it, build a billion-dollar company by themselves. But man, the fun of building a company is working with the people, the clients, the employee like that is the fun part of it, you know. I don't know that I want to build a billion-dollar company by myself. I mean, maybe, I mean, but for the money, okay. But I mean, is that the fun in building a building and building a company to do it by yourself? And so it's so what I'm saying is these narratives just keep driving this whole thing. Like you don't need anyone else, you don't need anyone else to build a unicorn, Nick. You don't need it, you don't need partners in venture to run a venture firm, just do it yourself, right? Right. I mean, and I think that's the stuff that's really, and I think you know, maybe for venture it makes sense, right? I don't know. But in the world of building a building, it makes no sense that somehow we're gonna like keep doubling down on these silos and do everything ourselves because I still think it's very much a human collaborative business. Now, to your point of like, well, it doesn't that mean that there's supposed to be a lot of data locked up. It already is, right? It already is locked up. I remember during BIM BIM days, we said there's gonna be a federated model. Get this. There's a federated model.

SPEAKER_01

I remember this.

SPEAKER_00

We're all gonna log in and design together. That went well, right? If you look at every construct of every every system, you know, so so let's let's think about the master builder, right? I keep, you know, one, it's hard to say master builder anymore because someone will get mad at me for using the word master. Got it. But what was really the master builder's job? Was it deep technical expertise in the early days? Yes, but it was really leadership, right? It was vision, leadership, understanding. Maybe not as much empathy back in the day. It was probably a little bit harsher, but understanding, right? I know what that carpenter needs to do, so I need to be thinking ahead of how this decision affects that person. So, you know, we've been playing around with alternate terms to the master builder, and the one I've been floating around is like the project success manager because people understand customer success. I don't love it, so email me if you have better ideas, I'll buy you dinner. But I think what what'd you call it? Pro the project success manager, yeah. Okay, because I think a project manager means you're managing a project, it does not imply that it'll be successful, but I think project success is really that everyone involved with the project wins. I feel like project management is I'm managing the project and I'm stating the facts. Oh, this is gonna be late. I mean, some of the worst project managers are just great communicators of facts. Well, this is gonna be late. Now the projects are gonna be late. I'm like, great. How can we not make sure it's not late? Well, it's just what it is. It's like some of these, you know, PMO, PMP folks, they're very focused on tracking the data. Same thing with like Trello boards and all this stuff that you know, what we call the keyboard workers and software, right? They don't do anything, they just track things. And so I think the difference is if you're gonna have a successful project, it requires a deep understanding of what everyone needs to do, great empathy to understand their point of view and and lead the project. So it's just more it's more about leadership than it is like I really have I have 50 years experience building buildings. It's really the people factor. And I think that's that's the issue. Yeah, that's the for that to me, that's the future, right? And I think I'm gonna experiment, you know, we have some deep, deep experienced construction management people on our team. I'm actually thinking about should I bring someone in that's with no experience but just has amazing leadership abilities where I can get people to take the hill without asking them, you know, they just you know the type of people on top, just people that are really good at that. Like, does it matter that they know how drywall is put together? I don't know.

SPEAKER_02

When you look at when you look at investing in startups over the next three to five years with all of this context in mind, how does that change what you're excited about, what you're looking

Investing In Founders Who Lead Markets

SPEAKER_02

for? I mean, less about the specific company and the problem they're solving, but more so like what are the high-level attributes that might have changed that you're looking for?

SPEAKER_00

Well, in my book from two years ago, right? I have a paragraph in there or a chapter, I can't remember. It's like the future with AI is that people no longer lead people, they lead markets. And I think you have to invest in people that understand big markets and understand how to create a leadership position in them. If you look at who's been very successful, right? Elon Musk, whatever you think about him, Alex Carp, whatever you think about him. I mean, Alex Carp doesn't write code, FYI, right? He's not Dr. Alex Carp because he's got a computer engineering degree, he's got a philosophy or a French degree or something. But he somehow figured it out, right? He got people to take the hill, got to understand, you know, he somehow took over ontology, which he doesn't own ontology. But but if you look at some of these folks, they never really focus. I mean, a lot of their people might even say that they're not great leaders of people. But what they've figured out is how to lead markets, how to understand market dynamics and really lead a market. And I think that's what it's really going to start to take. You know, we've always said that like founders try to solve problems and you know, sometimes they have to pivot and whatnot. I think right now it's really people that can absolutely lead markets. And I think some of these exits that are happening with AI companies, I think the reason they're selling is these founders, a lot of technical founders, nothing wrong with the technical founder. I don't think they've been able to lead a market. You have two people, two to three people, right, leading the AI market. And so I think it's either like sell and be part, you know, and be part of something with someone else, or you have to figure out how to be a leader. I mean, I mean, don't you think it's interesting? There's plenty of talented people, right? But yet it's Sam Altman, Elon Musk, Amade, you know, what's happening?

SPEAKER_02

Dario, yeah.

SPEAKER_00

Dario, right? Why are there only three voices? Maybe Mark Zuckerberg while he's doing, you know, taekwondo or whatever with his social media. Like, I don't know. Like, why aren't there more? Right? Jack Dorsey's a great example, right? What happened with Twitter? Why couldn't he get it together? Right. So I think the people you're seeing that are leading these markets, you know, people are saying it's a monopoly, but what's the what's been the difference in terms of them quote unquote being a monopoly? They have led the market, they have defined the markets, they've led the markets. Jensen over to NVIDIA, right? Does NVIDIA have the best of the best? I don't know. He seems to get people interested, he seems to get people engaged, right? The seal of AMD, she's fantastic. Don't see her running around, don't see people wanting to get her autograph, right? And so I think as we start thinking about how do you spot talent for investing, I think you're trying to spot talent around people that can lead markets, that they will be the top five leaders, maybe not from the tech, right? It may not be the best tech, it doesn't have to be, but that they can actually lead a market and have a voice that people listen to. Like, like I think like Ryan over at what's the shipping business? Go, it's a company.

SPEAKER_02

Shipping business.

SPEAKER_00

The three people they like do shipping and all that. Anyway, I forgot the company. But but you know, there's some of these guys that really do, like they they are leading markets.

SPEAKER_02

So it's always maybe more so the founder attributes are in this era, maybe still maybe, maybe just magnified. The leadership capabilities, the and the ability to lead people in markets is maybe the it's it's uh it's it's weighted higher than than what you would have weighted it previously in the attributes of a of a investment underwriting process. Yeah.

SPEAKER_00

And what do you think happened? Lovable, right? A lot of people love lovable. What's the CEO's name?

SPEAKER_02

Couldn't tell you.

SPEAKER_00

There you go. Right, there you go, right? I mean, so they just raised another round. I'm not they probably have a great product, right? But they're gonna they're getting washed out. You're gonna, I mean, I predict, right? You're gonna see a press release that comes up that says they sold to someone's gonna buy them. Yeah, yeah, they sold to anthropic, they sold to whoever.

SPEAKER_02

Yeah.

SPEAKER_00

So I think in these big waves, right? Not these like little waves, right? These big waves, I think that's what it's gonna start to take, right? Is these people that can really understand the market and know the psychology of it and know how to lead it. I mean, where's Mark Benny off in all this?

SPEAKER_02

He was the software guy.

SPEAKER_00

Yeah, but then you look at Brett Taylor. Brett Taylor's doing an amazing job.

SPEAKER_02

He is doing a great job.

SPEAKER_00

God is just doing such a good job. There's a reason, right? I mean, the guy is like talent, he's top right of the people out there. I think he's like top three talented in the world, and most people don't even know who he is, but the people that need to know do know who he is, right? You don't see him on CNBC jumping up and down on a chair, he just shows up and does the work, right?

SPEAKER_02

Yeah. I mean, I in in Silicon Valley, he's he's definitely recognized. I mean, he was one of the shortlisted people when Sam Altman was removed from his CEO position for a couple days. Yeah, Brett Taylor was on the I think he's on the board of OpenAI.

SPEAKER_00

He's the chairman, sorry, he's the chairman.

SPEAKER_02

He he like, yeah, there were text messages where he basically agreed to the position under certain terms, and he was yeah, maybe one of three people that was recommended. And I've even seen people yeah, recommend at even at this point in the juncture that open AI makes that change eventually.

SPEAKER_00

I think I think open AI is gonna buy Sierra, yeah. I think it's just gonna happen.

SPEAKER_02

But he's just yeah, he's got he's he's definitely got like he doesn't have the same charisma as an Elon or a Jensen, but he has the CEO attributes and man does he execute. And I think he's just got a great reputation.

SPEAKER_00

Yeah, I mean leadership takes all forms. I mean, I always look at it like if if you could have lunch with whoever you wanted to have lunch with, right? He'd be probably top two, yeah. Number two only being Warren Buffett because he might die soon. Like, there's a like I'll get a if I get a second shot of the magic lunch, right?

SPEAKER_02

Better get Warren Buffett out of the way, like you know, you put uh you put Brett Taylor over Alex Carp. Yeah, wow, yeah, surprised by that answer. I just I think you're exaggerating, but okay.

SPEAKER_00

No, I think Alex, I mean, I think would I like to spend time with Alex? Sure. But I think like Alex Scarp's a better hang. He's a better hang, but I think I'm gonna be like just listening him, listening to him talk.

SPEAKER_02

It's not gonna be very uh I don't think I'll be on a two-way conversation.

SPEAKER_00

Yeah, yeah. I mean, just ask every journalist a separate. But I but I think if you I think someone like Brett, if you said, Hey, what do you think the next five years looks like? I bet he has a very specific idea of what the next five years look like. Yeah. Without a lot of hyperbole, right? Like just here's what I think things are looking like.

SPEAKER_02

Yep.

SPEAKER_00

You know, and I think if I had you know, hey Brett, let's grab lunch. You know, in the near term, if if if there was someone I wanted to spend time with, I'd really want to pick his brain. Very cool. So all right, man. Well, that was a good episode. Good to have you back. It's good to be back. By the way, I am I am on vacation next week.

SPEAKER_02

Oh, yeah. Wow, okay.

SPEAKER_00

Allegedly. Well, what I've done is I've told the family, and

Vacation Plans And A Solo Pod Tease

SPEAKER_00

you you know this well, you don't really go on family vacation, you go create experiences with your family. It's not really vacation. So I've I've cleared the books of all internal meetings, non-essential meetings, and just only ex external extension, extension, uh external meetings that are essential. And I think we're down to like six.

SPEAKER_02

Okay.

SPEAKER_00

So only six, only six meetings on my vacation.

SPEAKER_02

The part-time vacation next week. Okay, okay. But all right. Well, our listen our listeners can um what sorry, what'd you say? Do you want to host host it without me next week? Oh, maybe I'll try the solo episode.

SPEAKER_00

I'll send you the link.

SPEAKER_02

That would be uh yeah, probably. I mean, it might be a two minute episode, but we'll see. We'll see. All right.

SPEAKER_00

All right, we'll see you then. See ya.