AI After Dark

00;00;00;07 - 00;00;23;07
Unknown
And I really wanted an answer. How does an AI native technology get adopted by a legacy industry? You know, how does a restaurant, how does a oil and gas company as a real estate, you know, company or whatever? Slowly. Maybe painfully? Yes. Yeah. Suddenly there's a breakthrough and then you go back three step. Well, and I think I think over time, yeah, over time it's been, and something I'll share with you after we're done recording.

00;00;23;07 - 00;00;44;18
Unknown
But, over time I'm realizing, like, the easiest way to get past the fear, uncertainty, doubt. Like FUD. It's just been like, don't like, pay me very little so that you actually have skin in the game. But let me just show you one thing and then you see the magic. And that one thing could be, by the way, could be like an artifact that Claude Co-Work creates.

00;00;44;18 - 00;01;15;24
Unknown
But it's one thing that then, starts the snowball, and it's like anybody that's created, build a snowball, try to roll. Snowball knows, like, the first packing sucks and trying to roll small ball sucks. But, like, once you get to a certain point, my bad. It's interesting to see, you know, I talked to a lot of people about how they're approaching bringing AI into their business and their company into their workflows, whatever it is.

00;01;15;25 - 00;01;51;06
Unknown
Yeah. And there's some and I like look at what a lot of people sort of doing what I'm doing right now are doing where they're like, what, one step out, what you want. I thought of it, so I I'll try to give the like, reasonably efficient version of how I got here. I, I'm currently running a single man solopreneur ship thing called Houston ai help which is doing AI implementation work kind of embedded in small and mid-sized firms, mostly here in Houston, not exclusively.

00;01;51;08 - 00;02;11;10
Unknown
But IRL, geographically can stream but born out of out of an afternoon of having margaritas with my wife at content way back in January, where I was like, I'm using these tools. So much, and I'm building things both for my business. I was running this dental. I am still I own this small business that sells dental benefits.

00;02;11;12 - 00;02;29;12
Unknown
And I started building tools in there, and then I started building stuff to run my little league baseball team. I started posting, like, on LinkedIn, like at this thing I built to draft my little league team, showed it to a guy at the co-working space I was working out of who coaches a softball team, and his reaction wasn't, can I have that for my softball team?

00;02;29;12 - 00;02;54;07
Unknown
It was like, oh, can you do this for like, lumber take off estimates in my homebuilding business, which was like that light bulb moment next afternoon, my wife and I go to have a margarita and some queso before we pick up the kids from school on Friday. I was like, I think, like, there's maybe something here. So you kind of like paired subject matter expertise with AI tech talent, basically.

00;02;54;07 - 00;03;13;25
Unknown
Yeah. My background, like I was at I was at Uber for seven years or early. Uber started doing operations and logistics in the DC market and like was not a I've always liked tech and software, but I'm not a software engineer. I took one computer science AP class way back in the day. But I've been around tech.

00;03;13;25 - 00;03;31;18
Unknown
I spent time in tech. I ended up in the product ops organization there, which was sort of product management adjacent. So I'd done some of that, and then I moved back to Houston. I buy a small business. I'm in this place where all this interesting energy stuff is happening. I just feel like I found myself in this funny, I've got a tech background, I've got a small business owner background.

00;03;31;18 - 00;03;49;20
Unknown
I've got, I've done a little bit of consulting as, like, all these things right now feel like they come together into this little thing and we're going to go see what happens this year. So it's been fun. And we can get into like, the stuff that I've been been building with it. But I think, like a lot of people doing this sort of, hey, let me come help you with AI.

00;03;49;20 - 00;04;16;15
Unknown
There's a couple different approaches to it. One of them is we're going to do this month long discovery process in your business. We're going to talk to all the executives and then go all the way down the organization. We'll come back with a map and like, here's everything. And what I have found for this, the businesses that I'm talking to around Houston, the size that they are, though, like they don't often have much of a tech, internal thing.

00;04;16;18 - 00;04;35;06
Unknown
They still don't really know what they don't know, and they know that I guess they know that there's a lot that they don't know about how I could impact their business. And the idea of paying someone for a month of discovery is like, not, I think, the easiest way in and a lot of cases. And so I try to find a few of those like quit.

00;04;35;13 - 00;04;56;28
Unknown
Sorry. And yeah, this is more ignorant, but does it take a month to find to discover everything? Or like, could you do it in a three hour whiteboarding session? That's more what I try to start with. Okay. And I guess I would say we are not going to find everything in a three hour whiteboard session. In fact,

00;04;57;00 - 00;05;14;22
Unknown
So so my, the client that I've now worked with for the longest, I'm sort of on like a indefinite retainer with them, which is a great. Yeah, great situation for both of us. I think it's working really well. I went in, I did two hours on a whiteboard with several people from different parts of there. Just the lumber.

00;05;14;26 - 00;05;38;10
Unknown
This is not the lumber. So the lumber guy, he and I still chat all the time. I've helped him a little bit with some stuff, but. But that's that's not him. This is a, this is the arc 57. They are a kind of a holding company, like, operating energy adjacent businesses for the long term, basically, like, entrepreneur in residence.

00;05;38;10 - 00;06;02;16
Unknown
Put them in a business, grow this thing. So they've got, like, a commercial generator repair business and they've got, a, they, they do pipe rental for drilling wells out in Midland, like they've got these companies that they own and operate, kind of by embedding operators in them, which is great because there's these there's both the like parent company and holdco strategy piece of this.

00;06;02;18 - 00;06;21;10
Unknown
And then there's the like individual operations that I'm sort of touching things and all. But we we start with this two hour whiteboarding session. There was just like I just enjoy the conversations right. This goes anywhere or not. Fine. Like I'll come chat about your business for two hours with anybody. I don't know if I should say that.

00;06;21;13 - 00;06;46;16
Unknown
I'll give you anyone two hours of free, free work to go and talk about it. But, and we started we started to identify some of the opportunities that could be there to do some work. But the map that I would draw today of how their business works is very different now, four, five months later, whatever it's been, it's very different than the one that I think I could have articulated coming out of a two hour whiteboarding session.

00;06;46;19 - 00;07;14;08
Unknown
But the decision that we made coming out of that was like, yes, let's try this for a few months, see if Zach can come in and help us build some stuff into into what we're doing. But it was even more like, we're going to get a couple quick wins. You're going to start to see how these things can impact your business, and that's going to help us start peeling back the layers in other parts so that we can do some of the bigger stuff.

00;07;14;11 - 00;07;34;13
Unknown
And because because I envision a situation where it's like as you start uncovering things, you realize there's much more problems and there's like an infrastructure issue before you even get to those quick wins. Yes. So what quick wins are you getting in the first couple weeks that are enough to be able to, like, give you time to fix all the back end?

00;07;34;16 - 00;08;03;27
Unknown
It's a really good question. So to use a very specific example, one of those one of those businesses has this field service platform that they run the commercial generator business out of. That is it's a software as a service that they use to go do all of their generator repair. And it's designed to be a closed environment, like you're supposed to be able to do everything in there.

00;08;04;00 - 00;08;25;25
Unknown
So you can't write data into it. There's no API that you can write into. You got to do it in their interface. But there is a, for, for analytic analytics purposes, they give you a way to get data out of it, read data out of it, and it's not live. It's kind of like a once a day snapshot replica thing, whatever.

00;08;25;27 - 00;08;57;09
Unknown
But the the ways that the team could run reports within the tool were, were pretty clunky. And they were they couldn't draw the connections that they wanted to draw. And when I realized sort of in my first week there, after we'd had the discovery session, like, okay, I can at least read all of this data, I think we can probably build something at least to, like, help you do the analytics that you haven't and give the president of this small business visibility into the way things are happening that he can't quite see.

00;08;57;09 - 00;09;01;26
Unknown
And he's able to articulate to me like, I think there's something.

00;09;01;28 - 00;09;31;28
Unknown
With how we're, you know, dispatching jobs or routing, I don't remember even the exact examples, but like, there's some data in here that I if I could see it the right way, it'd be really interesting. And when I realized I could get read access to this whole thing. Well, okay, I can't I can't help you, like, put stuff into it, but we can start creating you some tables and dashboards and graphs and things, you know, put together a little web app and, kind of the first day I'm onsite with them, at the end of the day, I'm like, all right, here's just an example.

00;09;31;28 - 00;09;53;03
Unknown
One page and he looks at it and goes, dude, you just didn't six hours. What we've been trying to do for six months, which was like the best thing I could possibly hear as a as a first day on the job, sort of consultant implementer guy who created a little bit of a like, okay, well, I hope there's still another several months of work that we can do on top of this.

00;09;53;05 - 00;10;11;23
Unknown
That obviously was, but yeah. Sorry. Continue. No, please jump in. Because I was going to say I, I was talking to somebody else and, and they're much more focused on the training side. Right. And they were like a former Google for meta. And and he's like, you know, I'm basically training people on how to use these different tools.

00;10;11;23 - 00;10;26;10
Unknown
And inevitably you end up with like, well, let's build a little this a little bit, a little that. And he's like, and so my job is basically become first of all last mile. Or I come in and I teach the first mile and then they work on stuff. And then I come in to make sure that, like, what they're building is scalable.

00;10;26;13 - 00;11;00;22
Unknown
And there's this last mile component where you think people think, oh, once it's done, it's done. The last mile never gets there. There is always more. There's always more things to do. Welcome to any business ever, I think. But, yeah. And it's, you know, I, I do some of that kind of more training type stuff as well, either with individuals who want to do that or some like I, we've been rolling out cloud for teams and across this organization and doing some of that, and I've got a I've got a lot of things I still want to do with them, but getting a team that's never used those tools up that curve, it's

00;11;00;22 - 00;11;26;27
Unknown
sort of the what's going to make the bigger impact faster. But then also what's the right long term investment? It's a there's a tension there. And I think it depends on the organization. Right. You're not going to have guys who are going to customer sites and repairing generators, talking to Claude or ChatGPT all day. I mean, I can envision a product that we could make them that might allow them to do some of that.

00;11;27;00 - 00;11;52;06
Unknown
But right now, the field service platform runs in this closed shop that can't connect to it. So we'd have to completely find a new field service platform and like it would require too much change. And I think maybe in the next 2 to 5 years that type of thing will make sense. But if if I go into a business and then the first thing I say is, okay, well, we're going to strip out all your software and build you brand new software that's going to have AI embedded in it and all.

00;11;52;06 - 00;12;16;17
Unknown
That sounds nice. It might be a decent pitch, but I know that one. It's going to be harder to do than like doable, but harder to do than anyone wants to be. The change of that is going to be really hard for the organization. And frankly, if it's the piece of software that, like, literally runs the business on a day to day basis, I'm a solo developer, product manager, implementer, strategist, whatever.

00;12;16;20 - 00;12;34;26
Unknown
I don't want to build anything that if it goes down for 24 hours, that business can't run like the. The worst moments in my days at Uber was when there was a system outage and like, rides could not happen. Like everyone's just sitting there going like, come on, we got to get the system back up. And there's on-call engineers who are doing all that.

00;12;34;26 - 00;13;14;02
Unknown
I don't have one guy. I am the on call engineer and I have kids to pick up from school. And yeah, so do you, how do you get around that? I mean, you're building a digital twin in testing there before you implement. Yeah. So I, so, so two things like one is if we're building, like, tools that help the, the sort of executives and managers in this business understand what's happening in the business, maybe like if, if we're building, something to, like, make their sales process more efficient, it's possible that that could go down and like, certainly inconvenience a day, but the business doesn't stop making money because this tool

00;13;14;02 - 00;13;34;23
Unknown
is not not there. And so I'm uncomfortable taking on that project. It's not that I don't want to impact things in the operation of the business, but if it's like a core, this needs multiple nines of uptime kind of thing. Not that I think I'm going to build things that are gonna break all the time, but we've all worked at, tech companies and software that shows us like it's going to break.

00;13;34;23 - 00;13;55;08
Unknown
Am I allowed to say shit on this podcast? Yeah. I've said worse is you've called me worse. So I, that's a delicate conversation, and it's one I'm willing to, you know, if we get if I get to a point with a client that hasn't happened yet, but it's like we want a custom thing to run this.

00;13;55;14 - 00;14;28;14
Unknown
We want you to build it. We can have the conversation about, okay, well, what additional resources are we going to need to get involved to make that something that is sustainable and like doesn't just work when it's going well, but also works with. Yeah, well, I envision you as like a fractional chief AI officer, right. Where and where and if that's if you're okay with that, then like your role is a constant assessment of what should be built, what should be bought, what should be borrowed, you know, 100, and which, by the way, is a complete full time job.

00;14;28;16 - 00;14;54;25
Unknown
Can be I mean, just keeping up with everything, right? Like, and that's like, I, I, I think everybody also has a little bit of analysis paralysis again, because they're trapped in this FOMO and FUD environment and, and still don't really know how to calculate ROI with some stuff. And so there's all this new amazing tech and they're not, you know, it's a whole nother thing, right?

00;14;54;25 - 00;15;16;09
Unknown
Like an explosion of entrepreneurship. Because now people are able to visualize a lot more clearly what they've been trying to articulate. And now if you can do that, you can probably get someone to at least build it for you. Build it well for you. Right? You can definitely vibe cut it yourself. And, and so now, like, as a buyer, you're sitting there like, what do I do?

00;15;16;10 - 00;15;37;02
Unknown
You know, Zach, please, I trust you. Tell me what to do. Yeah. And I think my my first answer is we're going to take the time to figure it out together and build some of those quick, like, quick wins along the way. And I'll tell you some of the, my parameters, like, I'm not going to come in and rip out a core piece of your software and rebuild it today.

00;15;37;02 - 00;15;58;28
Unknown
We might get there, but that's not where we're going to start, which I think comforts a lot of people. I think the other tricky thing that I found, so in parallel to running this business, I also now run my own entire business essentially out of my own. I it actually works, whether it's in Cloud Code or ChatGPT Codex or whatever.

00;15;58;28 - 00;16;29;27
Unknown
But like, I've built my own internal system that I'm running all this stuff on. And, sorry, when you said this stuff. What is that, this stuff. Yeah, I am so, when I am working, when I'm at my computer, I am probably doing 80% of what I do in interaction with agents on my computer. So, like, it's, you know, I have a task management system, but it's connected by API to my agents.

00;16;29;27 - 00;16;56;29
Unknown
And like, I've got a skill that I built called What's Next based off of the West Wing TV show, or President Bartlet would say, like, all right, what's next? I can do that with my agent. And it looks at my tasks and it looks at my calendar, and it knows the shape of my day because I have this morning triage thing that I do, and it can say, all right, here's the next highest priority task, and it can pull the thing and chip off a new, a new agent to start that conversation, and we'll start working on it.

00;16;57;02 - 00;17;07;03
Unknown
So my my system to do everything is right now is basically built in these, in these tools. And.

00;17;07;05 - 00;17;31;09
Unknown
So like I, I in using this system see stuff all day, every day that, you know, I'm scrolling LinkedIn or x Twitter or whatever. And I see like I'm using 35 consecutive 35 agents at a time to like, do all of these. And it's all in the cloud. And I've got grok bot doing this and we have Hermes over here doing that.

00;17;31;11 - 00;17;50;20
Unknown
And I feel the fear, uncertainty and doubt of like I'm running like three agents locally on my system and I've got clients paying me for it. And I'm like telling them that I'm the expert. But these guys are all doing whatever. And I'm still pretty confident. There's a lot of hype and a lot of that. But like, it is sort of my job to experiment with these things and try things.

00;17;50;20 - 00;18;15;29
Unknown
And to some extent, that's how I've ended up where I am. I forgot how I was about to tie this to, to what I'm doing for my clients. Well, you were learning. You were. You're doing a lot of this in house, and which I assume allows for you to not just educate yourself on how to build it, but also identify, hey, where are the gaps that I need to go out and source externally and I can't just build myself?

00;18;16;00 - 00;18;34;27
Unknown
Yeah, I think that's right. And so like, I've been building a lot of stuff for myself, but then like, I still use a tick. Tick is the task management system that I use and I pay subscription for for years, and it works really well for me. And I could rebuild it, but it works well and it's got an API and it's however many, not a ton of dollars a month.

00;18;34;27 - 00;18;56;08
Unknown
Like I'm not going to go rip that out. Like let's let's keep using it. And so I can do that. Like we can have that conversation with my clients as well in terms of, all right, the tools that you're using are not necessarily the wrong tools, but what I, where I see those like light bulbs go off for people is when the different parts of their business and the software that runs it don't connect like that.

00;18;56;11 - 00;19;11;21
Unknown
There's no way to get the data from one into another, or a lot of the data ends up like we copy a little bit from here and a little bit from here, and it goes in Excel file, and we all email it to each other and we look at it, it's like, okay, we can we can make that process faster.

00;19;11;21 - 00;19;33;04
Unknown
Less manual data can get stored somewhere where in two months we can easily look back at what we did and how it's changed over time. And then they're able to make better decisions today and in the long or faster decisions and better decisions today. And I think in the long run, they're building kind of the the history of their business in a way that will it's like permanent corporate memory.

00;19;33;04 - 00;20;01;06
Unknown
I know. I mean, it's one of the things that I think I love, I love data, I love to look at stuff, run queries, whatever. When I went to Uber in 2013, I think it was the light bulb for me was like, oh my gosh, they have all of the data about the trips, about the requests, about who opened their phone, at what spot, at what time to look at what products and how many cars did it see, and what was the shortest data, and did they request and did the first driver accept the request?

00;20;01;06 - 00;20;34;26
Unknown
And like, I don't think a lot of businesses in 2013 had that level of of data available to understand what was happening in their business. But I sort of grew up professionally within that organization, doing those types of things. And so, like we we could run experiments at Uber where we could try offering an incentive to drivers on a Friday night to get them to be in a certain part of DC where we knew the bars were letting out and we'd get up Saturday morning at the time and and say, like, all right, did it work?

00;20;34;26 - 00;20;51;11
Unknown
And we could like literally tell from the data whether the thing we did last night worked, and it was going to be hacky and weird because the engineers hadn't built anything for, you know, ops folks were running queries and manually texting drivers and doing whatever. And then we would say, okay, this works. And we run a little bit longer and really make sure it works.

00;20;51;11 - 00;21;13;04
Unknown
And then we'd pass that off to a product team and say, hey, there's a way to actually influence the behavior that we want to see in this business and like help us build it. And I can now run that whole process myself, basically for my business, for my clients business, because I don't need a product manager and a designer and three engineers to build a product.

00;21;13;06 - 00;21;42;17
Unknown
Being called code can get most of the way there, all the way there in a lot of cases. Yeah. Your vibe coding before vibe coding was a thing. Yeah. Yeah, basically. Well, and I think, I think that's likely what has to be attractive to a small business owner about you specifically, is that you can kind of bridge both worlds, a little bit of, hey, here's what's going to actually add value to your business, and I'm going to help you build it, or I'm going to even maybe even build it myself to a certain degree, and implement it and do the feedback loops, you know, versus this like dichotomy of I'm a tech guy, give

00;21;42;17 - 00;21;57;06
Unknown
me a problem, I'll build your product. I don't care if you use it or not. I'm just know that, like, I can build it. It's much more fun for me when I can come alongside them. And like, I love learning a new business. And I mean, part of why I'm enjoying doing this stuff with AI right now is I'm learning new stuff every day while I do it.

00;21;57;06 - 00;22;13;01
Unknown
It feels like fun. It feels like play. And when I get to sit down with a new client and in either the two hour discovery session or over the meetings we have over the first month, like learn more of their business. And now I have ways to record those. I don't just have to take my notes and try to remember them.

00;22;13;01 - 00;22;32;09
Unknown
Like now I just granola and record a conversation or transcribe a conversation. And then I take that transcript back to my little system, and I've got a debrief skill that I've developed that's like, here's how we get all the stuff out of it, and here's what the next steps are. And I don't know, just it's this there's a loop to it that is loops kind of a buzzword right now.

00;22;32;09 - 00;22;49;06
Unknown
But but but there is a like I can I can keep all of that in and help it grow and multiply on itself in a way that is fun and valuable. Yeah. That's cool. Yeah. Yeah. Well, and it's not just constrained to you. I mean, I know you're woman band right now, but it's not just constrained that customer, let's say.

00;22;49;06 - 00;23;08;18
Unknown
Right. Like you, it it almost feels like you can build modular lead because you've got, you've built these loops that are recursive. And, and then when you look at a different client or a different business within this private equity holding group, or she's me, I don't know if I could or not, but holding group, and be able to say, okay, well, there's some pattern matching here that we can just leverage what I've built over here.

00;23;08;21 - 00;23;26;09
Unknown
It's awesome. It's fun. Yeah. Dude, that's that's why. So, I because because I remember talking to, you know, a guy a long time ago, and it mostly because, you know, he was like, hey, you're a VC like I what's up? You know, and you'd kind of share a little bit like okay cool. Can you, you teach me this and like, yeah, sure.

00;23;26;10 - 00;23;42;08
Unknown
Like, here's how I would do it. My job. Oh okay. Like like, can you teach my team? Yeah, sure. Like, you know, and and and being like, dude, you should you should go into like, all the mom and pops and do this, you know, and like, everybody's desperate for someone to walk in the door and be like, let me show you how to use AI.

00;23;42;11 - 00;24;02;04
Unknown
And I said, yeah, but if I did that, they'd be like, get the fuck out of my shop. You know, like I who are you? So my question to you is like, how is that growth or go to market journey been. Yeah, it's an. There's definitely the occasional I mean I'm not a big like cold call or go walk into a mom and pop.

00;24;02;10 - 00;24;20;11
Unknown
So so I guess I don't, I don't get go the fuck away very often, but but there are a lot of people who are sort of seeing the same wave that we're all seeing and saying, I can go be the guy that does this stuff. I think,

00;24;20;14 - 00;24;52;07
Unknown
The go to market is interesting because it requires both, my skills to be right, but it also requires a business that's actually ready to, to do this and there's clear value to be had right now. But also I still think I mean, I think it's pretty clear we're very early and what all of this is going to do and so, you know, this really cool system that I've built right now or the software that I built into your business very well may be obsolete in 18 months.

00;24;52;13 - 00;25;13;19
Unknown
And there are a lot of businesses that don't care to be that far on the cutting edge, like they may be monitoring it and feeling like they're missing out on something to a certain extent. But they also see the security issues that get a lot of headlines, or the way a product changed in three months or whatever. Like I don't they don't want to be running their business in a way that deals with that.

00;25;13;21 - 00;25;46;06
Unknown
And so there is a, there's a fit that is required there where, where a business needs to be the type of business that wants to be on the cutting edge of that kind of thing and be willing to absorb, I will do everything I can to soften, soften the pain or the challenge of, of that transition. But I, I can't know for sure that something's not going to change next month that will require us to rebuild three things, or that would benefit us from to rebuild three things that we've already done and that, frankly, maybe if they'd waited six months to do it, they could have done it.

00;25;46;09 - 00;26;08;18
Unknown
We could have done it faster and easier and cheaper then we I can't, I can't no. So, I guess I'm, I'm looking for customers that. See things similarly or exploring things similarly to the way I am. And that sort of makes it easier to not be. I need every mom and pop shop to to say yes to this, because I don't really want to convince people they need this.

00;26;08;18 - 00;26;24;04
Unknown
What I would prefer to do is explain to people how I do what I do, what I'm trying to do, how I'm thinking about this space. And if that sounds good to you, if that sounds similar to how you're thinking about it, and you just don't have the expertise or the bandwidth to do it, well, then let's have a conversation.

00;26;24;05 - 00;26;46;15
Unknown
We can go from there. And there have been a couple people who kind of, I think maybe we want to work together, but we're not. It's not that I'm like firing customers, but it's like, I just don't think this is we're not quite aligned enough for it to make sense to go into a big project together. On do you feel like they have to have an entrepreneurial, lens of the world?

00;26;46;22 - 00;27;18;16
Unknown
Like just thinking. I mean, most mom and pops likely are being run by the entrepreneur that started them, but there are definitely situations where there's a professional CEO or or like the kids took over and they're just not those entrepreneurial and and maybe they're looking at it to your point is like, hey, what efficiencies can I gain and give me some point solutions that can solve that versus, no, I'm on this journey to create this like eye for forward I native environment, that allows for me to do significantly more than what I'm doing today, not just ripping and replacing.

00;27;18;18 - 00;27;46;24
Unknown
I think it's less about an entrepreneurial mindset and more about whether a business is whether their medium term outlook is like pretty growth oriented or not. You know, I think if if you're running a company, it could be a pretty big company, but it, it operates at a certain scale with a certain team, and it's profitable and it's doing well.

00;27;46;24 - 00;28;22;24
Unknown
And, you're the business owner and you've got the setup, the, the life flexibility that you want, whatever it is. Then bringing in the challenge of innovating in this brand new space, it's just sort of to what end are we are we doing that? Versus an organization that's trying to grow new businesses or is is trying to expand their business, or feels like we've got, a team that we like, but our margins are a little low and we feel like we could do more, but we don't quite have the bandwidth.

00;28;22;24 - 00;28;41;21
Unknown
Well, that's then where these. Yeah, I rarely am saying, let's implement implement AI and get rid of people on your team. Like I'm not trying to replace functions. It's more of a how can we use this technology as a lever to help your existing team do more? And if you're not really trying to do more, you don't see the market opportunity to really grow.

00;28;41;21 - 00;29;01;18
Unknown
Then. So, you know, what do you care? That's interesting. And so it's almost like the whiteboarding session is, is a triage mechanism to understand where the mentality is like, yeah, that's right. I, my wife, she never listens. So this is easy for me to just to rag on her. But she, you know, she, she, as we've had kids, is, you know, it didn't she was a teacher.

00;29;01;18 - 00;29;16;11
Unknown
It was like, all right, it's too hard to do full time, but, like, doing some part time here and there for different things. One of which is for her dad, who's an attorney and has a law practice with her brother in law and so she kind of do more legal admin type stuff, right? And she complained about this, that are there.

00;29;16;11 - 00;29;35;16
Unknown
And I'm like, that sounds like it could be solved by AI, right. And and of course, like the first mistake I did was trying to solve the problem. I said, yeah. So next time I listened and, you know, you know, provided comfort. But after the fifth or sixth time, I was like, I can't we can't even keep having the same conversation.

00;29;35;16 - 00;29;57;00
Unknown
Right. We need to like, but, like, can I please teach you how to do this? Right. And, and and this is, by the way, this has been going on for a year. Last week, she was like, all right, let's sit down. Right. And so we go through her workflows and, and, you know, yesterday she's like, you you've completely automated my job.

00;29;57;02 - 00;30;13;02
Unknown
And I was like, that's great. Like, this is us versus what do you do now? And she's like, I don't know. I'm freaking out, you know, and and and it to me it was like such like cold water moment right where it was like I, you know, truth is the problem really. Was it this fear of like, what do I do now?

00;30;13;02 - 00;30;38;27
Unknown
And if you're not already thinking about that, if you're not the type of person that's seeking something like that, that is a complete detractor towards being someone that would hire a Zach to come help out right. And, and, and it's it was I mean, maybe I was too close to the problem and so I didn't tease it out or, you know, but but I, I have to imagine like, for someone that is going is trying to understand whether they're ready for AI or not.

00;30;38;27 - 00;31;04;16
Unknown
Like, that's a huge attribute. It is. It's funny because I've my wife left ten years in consulting at Deloitte to go on her own entrepreneurial adventure. Late last year, she started Houston Kids Yoga. Okay. She does kids yoga camps, and she teaches in a couple schools. And, does these day camps and stuff like that, crushing it.

00;31;04;18 - 00;31;34;13
Unknown
I built her a website and a booking engine and all that for her camps. And then, she sees me like, in the corner whispering to my, my whisper flow. Yeah. Spoken. Lee is my current default, so. Okay, so I run the local models and spoken like, but, she starts doing a little bit of that, and then she's like, okay, you've got this whole thing where you're managing your tasks and it's connected to your emails and like, can I have a little bit of that?

00;31;34;13 - 00;32;01;17
Unknown
And so it's been a fun least some of the traction is. Exactly. So, you know, the good thing, I guess, in her case is, doing mindfulness and yoga for kids is probably not something that the AI is going to, to replace anytime soon. And so I think in her case, it's a great example of if the technology can help her run the administrative side of her business more efficiently and she can spend more time doing the stuff that she actually likes to do.

00;32;01;19 - 00;32;24;10
Unknown
And that is where she delivers value. It's kind of weird to talk about a yoga and mindfulness in this business in the same breath where I talk about delivering value. But here we are. So, she probably will know, but but, I mean, I don't know, I it makes sense, I think personality wise, and maybe what we're describing actually is just like the journey that people are going on.

00;32;24;13 - 00;32;55;14
Unknown
Yeah. And like recently. Right, Sam Altman coming out and being like, you know, I, I just completely underestimated how slow and, you know, I don't know how like the macro economy functions. Right. And and maybe by design. Right. It's not quick to adopt. It's been fascinating. You know, there was a there was a clear. Well it's it's been funny for me because I still have a foot in the door of Bay Area Tech.

00;32;55;17 - 00;33;29;05
Unknown
But I live here in Houston, and the businesses that I interact with are energy businesses and, The, the conversations that are happening in venture back Bay area tech are very different right now than the conversations I feel like happening almost anywhere else. And there was a brief moment, kind of beginning at late into last year, beginning of this year, where there was a little bit of, I felt like the AI tools were making enough sudden strides to get everyone's attention a little bit.

00;33;29;08 - 00;34;04;11
Unknown
But then, probably not terribly surprisingly, that's led to a little bit of this cultural backlash of like, this is a lot of change really quickly. And what's it going to do to jobs and what's it going to do to our communities? And those are all really fair questions. And I don't it doesn't surprise me that those are the conversations that are happening, but it does it does create an interesting space where the I mean, I'm not unique for saying this, but the echo chamber, the bubble of the conversations that's happening around the cutting edge technology versus how everyone else is feeling about it and how it's scaring them for for what's happening in

00;34;04;11 - 00;34;38;24
Unknown
their job. I mean, I feel fortunate, I guess, to be at a spot in my career where I've done enough things and developed enough skills. But with enough runway left, where a new piece of technology right now feels like an incredible opportunity for me, right? Like, I can take all of the stuff that I've learned, and I've got a lot of runway to ride this wave, and who knows what's going to happen over the next 20 years if I'm coming out of college trying to get an entry level job and and businesses are trying to figure out how to automate a lot of the entry level work, like, I don't know how you learn enough

00;34;38;24 - 00;35;00;25
Unknown
to get to where I am today, right within yourself. Yeah. And so, and if you're later in a career and you thought, I've got another 5 to 10 good years before, you know, then I'm going to retire and be fine. And suddenly the job that you were sort of coasting towards retirement on is changing under your feet and might go away.

00;35;00;25 - 00;35;24;08
Unknown
Like, that's probably pretty scary too. And I, I'm sympathetic to that. I try to, I, I guess I'm obviously very bullish on the technology and the impact that it can have. And I, I tend to look you can look at, sort of all of human history. And there has always been the backlash to pretty much every new technology in terms of this is going to ruin the kids.

00;35;24;08 - 00;35;52;00
Unknown
This is going to ruin society. We've seen have you heard, I think it was Marc Andreessen on the bicycle face. Yes, yes. There was, a podcast that I listened to. I think it's dormant now. Ended up being called build for tomorrow, but it was largely based on looking back at how everyone reacted to the fork was, was apparently a really disruptive new technology at one point, like the fascinating thing to study.

00;35;52;02 - 00;36;08;22
Unknown
So I tend to my, my perspective is like, yes, this is going to be painful and weird. And also over the course of humanity, we we have largely been pretty good at figuring out how to incorporate new technology in a way that we all benefit from and feel good about in the long run. And I tend to think that that will happen here as well.

00;36;08;24 - 00;36;29;09
Unknown
But that doesn't mean it won't be painful as it happens. And this one is moving faster and more disruptive than even any other technology we've seen probably in our lifetime. And I, I understand that that's challenging. So this brings up a good point, right? Which is like, and I know you're not in the business is necessarily like, hey, I want to have a retainer forever, but it wouldn't be bad.

00;36;29;11 - 00;36;49;28
Unknown
And so, like, so to your point of at a certain, you know, there's going to be a moment where the broader society is going to be have adopted a lot of these, you know, foundational models or have at least learned how to use them and leverage them. What does Houston AI help look like a year from now, two years from now to where it's at today?

00;36;50;00 - 00;37;05;01
Unknown
Part of my answer to that question is, I don't know, and I'm not going to worry about it too much today. It's a good answer. And I it's fun to do the thought exercise.

00;37;05;03 - 00;37;22;03
Unknown
Sorry. It just started pouring. Raining. Now, we didn't when they. When they rented this place, they didn't realize that the metal roof would cause this kind of. Yeah, I like it. But there have been meetings where people have had to like, stand up next to the TV to be able to hear because it. Yes, Houston needs the rain, from it.

00;37;22;05 - 00;37;42;02
Unknown
So we'll take it. You might fall asleep by the end of this week. Yeah, it's it's a nice it's a nice, the microphone, but I won't pick up any of it. And it'll be, it'll be. What are they talking about? So, so to some extent, I don't know, part of part of why I'm doing what I'm doing, and I'm doing it by myself for now.

00;37;42;02 - 00;38;13;12
Unknown
Is that feeling that, like, I could have chosen to use the skills that I've been developing to try to, like, build a software product with AI or for AI or whatever, and it is hard for me to think about, like, I would be scared right now, frankly, to be raising venture money to build a software company based on a belief that what I'm building is going to still be adding value.

00;38;13;15 - 00;38;39;17
Unknown
Ten years from now? Five years. What I don't like. Things are changing really fast. Yeah. And I respect the people who are willing to make that bet right now. It's not that I'm saying that they're idiots or that they're wrong. It's just like, for where I was looking for the risk appetite that I had, I don't want to be taking a bunch of someone else's money to go try to build a unicorn.

00;38;39;20 - 00;39;03;11
Unknown
And because I don't, I don't know that I've got an idea that I might have some good ideas, but I'm not sure that in a year from now, the frontier models won't be able to just do the thing that I was building a piece of software to do. I think a lot of businesses are having to pivot through that challenge right now, or are struggling because these models are able to do that.

00;39;03;14 - 00;39;25;24
Unknown
I'm sure that they're going to be some incredible software businesses that come out of this era of things. I'm not I'm not the type of person who thinks that we're all going to be vybe coding every solution to every problem that we have software wise. But I do think it's going to be more challenging, and I that just wasn't the bet I wanted to take.

00;39;25;26 - 00;39;53;25
Unknown
So what I thought would be more aligned with how I was feeling about things is I'm having fun building these tools that are built specifically for my business and what I am trying to do right now, if I if I spend 90 minutes building an app to help me draft my Little League baseball team. That's reasonable. Even if I only use it one time to draft one team, I never manage again because it was fun.

00;39;53;25 - 00;40;11;25
Unknown
I learned something, and I've got a good tool to draft the team and came up a little short in the playoffs, but it's good. Yeah. Yeah, I don't blame the coaching, not the draft. And technical debt isn't that costly anymore. Correct. So just so just build it and see what see what happens. And I guess you could say that about the bigger picture idea too.

00;40;11;25 - 00;40;48;26
Unknown
But but what I loved about that was, you know, there was no off the shelf piece of software that was going to map my brain process for how I wanted to draft my baseball team into an app that I could use in real time. At that meeting with the other dads to draft the team. And it's a stupid it's a, it's a it's a silly example and intentionally so, but I think, I think there is an opportunity for businesses that can build those types of like, you're not building your process to fit a software tool that you bought off the shelf, which is what a lot of small and mid-sized businesses that don't have

00;40;48;26 - 00;41;07;24
Unknown
a tech component do. You can decide what you want your process to be, and you can build the software that helps you do that as effectively as possible. That's interesting to me. That's fun. That solves a problem for you and delivers value for you immediately. And it doesn't require solving everyone's problem for a very long period of time.

00;41;07;26 - 00;41;32;04
Unknown
Depending on what we're doing, it can be a one time solution. You can have an employee departing your business and, realize they've probably got a lot of knowledge in their email inbox that you wanted to have, like, oh, well, let's download that whole corpus of knowledge and have an agent roll it and have it ask you some questions about what's important in here and how it should be organized.

00;41;32;04 - 00;41;48;28
Unknown
And then they'll run off for a couple of hours and come back and you'll have this incredible exit package for the the person who left that required very little work from anybody. And maybe you won't ever need to use it again, but it wasn't that hard to build and you got a ton of value out of it. I did that for a client like it was a it was a fun little yeah project.

00;41;48;28 - 00;42;06;14
Unknown
I think that kind of thing might be disposable. It may literally may never get used again. And like it's still delivered value. I get it. Like I'm just putting myself in the side of the client here. Like this is so confusing, right? Like this, I mean, and I don't mean this in any kind of this way. I just mean like, please, you're telling me I go to market?

00;42;06;17 - 00;42;23;27
Unknown
It's great. Because, like you say, okay, hey, you can build this. You can just build things. You can just do stuff, right? You can just build things, and it can be one off use cases, and that's okay. Like all there's like a, like a litany of things that you don't do because they're like, well, how much value is there really to doing that.

00;42;23;27 - 00;42;46;17
Unknown
Right. And now you can and on the, on the other side, and maybe this is the prioritization effort that that you're like managing expectations through. But like on the other side you say, well I like there are things that in the range of profitability and comfortability like you don't need to build everything. Like you can just buy certain things because they're useful and they work and it's it's fine.

00;42;46;17 - 00;43;05;03
Unknown
It's just convenient. Right. And then so, so as a business owner, I'm, I'm, let's look at, look at all my workflows. Let's, let's say I get to the point where I can I've mapped all my workflows and maybe even figured out which ones are most valuable or not. How do I then decide what should I build versus what should I just like, you know, buy?

00;43;05;07 - 00;43;26;10
Unknown
Yeah. I guess I would say a couple things. One, I don't know that there is one answer to that question. And it actually leads me to a bit of a I want to talk a little bit about how we how you can use AI to help you figure out your own thinking about something and not just have it sort of do the thinking for you.

00;43;26;10 - 00;43;33;23
Unknown
I'm going to come back to that in a second. I think two things. One, I think.

00;43;33;25 - 00;43;56;00
Unknown
It's interesting to me how often the way like a workflow might get described. Well, one, how often the workflows haven't really ever been described. So even just the process of documenting what the workflows are tends to be useful, to how like the ideal version of how workflow works is often very different from the compromises that get made in the actual execution of it day to day.

00;43;56;06 - 00;44;25;10
Unknown
And I think sometimes the more, the more immediately impactful things are from helping support or minimize some of those compromises that have to get made in a workflow. But some of it's feasibility to like, hey, you know, we could try to take the hardest thing on your list, the thing that would be most impactful to your business, but it's going to take us a couple months to get to something that we think you could actually put into production.

00;44;25;10 - 00;44;46;00
Unknown
Or there's three things that five people on your team are using twice a week for a couple of hours like this. Could this this small process, we could have at least a first working version of it next week. And several people on your team might get a couple of hours of time back if it works pretty well to me.

00;44;46;00 - 00;45;04;28
Unknown
Like, let's start there. Rather than kind of the boil the ocean approach of like, what's the biggest meatiest thing? Because particularly if it's a, it's a client that we're just getting started with, this will help me understand how your team works. It'll help you understand how I work to start to see how these tools can can be impactful.

00;45;05;01 - 00;45;33;01
Unknown
And you'll start coming up with other ideas, like a lot of times, I was talking to to someone and I, one of my clients a couple weeks ago where there was this report that they got emailed to them every day, and then they were copying a couple columns into the spreadsheet each day that at the end of the week, they were sending out this other report, and there was a problem with the front end thing that emailed this report every day.

00;45;33;03 - 00;45;52;12
Unknown
And I'm having this conversation with her, and I kind of get she kind of gets to the end of explaining what she uses it for. And I said, so would it actually be easier if, rather than you getting this email every day that you copy and paste over here, like, what if we just built the tool that gives you the thing at the end of the week, which is it just hadn't.

00;45;52;14 - 00;46;13;28
Unknown
And I don't blame anyone for this because they're in that they're they're down in the job of doing it. They've got a job to do. I need to get to this spreadsheet. This report is how I get to the spreadsheet. And no one's ever been able to sit with them to kind of zoom out a level and say, well, if the goal is over here, sure, I can help you with this, but like, let's add the three more steps and automate them in a way that you get the final report.

00;46;14;02 - 00;46;33;27
Unknown
This is the one that actually matters for you in the business without you having to spend the time here, and then you have more time to analyze the output of this spreadsheet. That is what you actually ultimately want to do or to focus on the other project that you've been putting off for months because it hasn't been done yet.

00;46;33;27 - 00;46;53;25
Unknown
So I actually I love the quick wins, the quicker wins, because I think those actually sometimes unlock more faster than. Yeah, the thing that maybe feels. And if we wait a little bit longer on this and the technology gets a little bit better, it gets easier and faster to do this once we're ready for it to. So I, I actually start small.

00;46;53;25 - 00;47;26;21
Unknown
And so how do you surface that like talk to people. Yeah. Yeah. And so like that's that's when you say when I talk to everyone in the organization, that's where we really start. Yeah. And it's not, it's not necessarily, like I need not make sense. It's true. Yeah. It's. Yeah. Well it's it's it is to me it's like the, the fun tension of sort of this moment in technology of in some ways this is technology that can make us never have to talk to anyone again because it can just we can run the whole business as one person or whatever.

00;47;26;21 - 00;47;55;25
Unknown
But, I don't I don't think. I don't think you can figure things out. Just behind the screen looking at the data. I mean, I love I, I have tried really hard to figure out as much as I can, just like go back to the Uber story of all the data that we had. But frankly, like a lot of what I learned in that driver operations role was going out to, an airport parking lot and talking to drivers who were waiting for a ride or taking a couple taxi drivers to lunch at the Ethiopian restaurant in DC.

00;47;55;25 - 00;48;14;05
Unknown
That was awesome. And learning what was actually impacting the forks there either. So that's true. Yeah. And generally speaking, that was delicious. I loved it. So the people like you talk to people is basically the answer. And that's where I get to put on that kind of more product manager hat. And, just try to ask good questions.

00;48;14;07 - 00;48;50;13
Unknown
Yeah. Which I will, I will use to say that's one of my favorite ways to use AI too, is to get it to ask me questions. I'll give a shout out to Matt Pocock, British guy who is a talented software engineer who's gone into this space and has developed a set of skills, probably the most well known of which is called Grill Me, which basically is a skill that you can use in your systems to have it ask you a series of questions about something you're trying to figure out to basically sharpen your own thinking, make sure that you and the agent have the same thinking about it, and then, oh dude,

00;48;50;13 - 00;49;08;15
Unknown
I bet you. If not, I mean, maybe half, maybe not. I don't know, of the value of the work that you do is just sitting with people and forcing them to tell you about what they're doing. And when they start articulating, realizing, oh, well, why do I do that that way? What's the purpose of this? Well, maybe I should do it.

00;49;08;15 - 00;49;34;15
Unknown
This, you know, like and and and it's not even, you know, I quote unquote related. No, it's it is the fact that I can help becomes the forcing function that makes you document your processes and realize what you do. And so the thing that you thought you wanted me to come in and do, we'll probably do some version of it, but then we're gonna in the process of figuring out those, those aspects of what you do that you didn't even realize we're going to build the roadmap of all day.

00;49;34;15 - 00;49;47;14
Unknown
Yeah, yeah, yeah. Okay. So you don't you don't know what it looks like a year from now, but what do you want to do? I want to look like that's a fair question. I did a good job avoiding getting there.

00;49;47;16 - 00;49;55;13
Unknown
I am trying to hold that loosely for a number of reasons. I can imagine.

00;49;55;16 - 00;50;20;07
Unknown
That I'm still just doing this solo for 1 to 4 clients at a time, where, I get to get in there and get to know the business and do what I'm doing, I could see it growing to where I need or want a team to help do it. And more places or faster or be able to build things that do require uptime.

00;50;20;09 - 00;50;44;06
Unknown
Better than a 24 hour lag. So that that's like, that is a possibility. I could end up I'm not unwilling to I've not had a real job. And since I got laid off at Uber in the pandemic, and it's hard. I think once you go this long without a job, it's kind of hard to, like, get up today.

00;50;44;06 - 00;51;01;02
Unknown
I. This is a big thing. That's a big day. That's right. I try to look like I could go play golf at any time, because that is sort of like on any day where there's a good. It's it's. Yeah, it's professional enough. And I could actually head to the driving range if, if the opportunity presented itself.

00;51;01;05 - 00;51;22;09
Unknown
But but for the right opportunity and the right business that, that understands that, you know, picking up my kids at school is still important to me. I could see myself, in a in a full time role. Maybe, maybe. Yeah, maybe it gets to the point where it's like, okay, I now I've done enough of this.

00;51;22;09 - 00;51;37;09
Unknown
I see what the software product is or some other product is, and and it does become something that is not sort of service based but is more or product based. It's an interesting,

00;51;37;11 - 00;52;12;00
Unknown
I think the challenging thing is. That this business can't, doesn't exist unless I'm, bringing myself to it pretty fully. And, I'm really enjoying doing that right now, but I don't know if I want. I've never done anything in my life for more than a few years at the time. But really, I mean, even even at Uber, I felt like my job was to find a problem, solve it well enough that it makes sense to build a product around it, pass it off, and go find the next problem to solve.

00;52;12;02 - 00;52;31;15
Unknown
And there aren't many things that I've ever done for more than a couple of years before I start to feel that itch. And sometimes that that's an internal growth within the thing that I'm doing. And sometimes that's a a more significant pivot. So funny, like everybody ever recorded with has been like, all right, I want to come back in six months or I want to come back in a year, and I want to record again.

00;52;31;15 - 00;52;51;07
Unknown
Right. And it's always been granted mostly it's like CTOs and technical co-founders, whatever. And they're like, I want to be able to talk about how much things have changed, right? You're the first person that I feel like we should record a year from now, but mostly because, like, how the business in itself has changed of what you do.

00;52;51;07 - 00;53;08;29
Unknown
Yeah. I, I'm happy to I'm happy to go ahead and pencil in it. I really don't know what we're going to find. And, but I think that's part of the fun. Yeah. That's here. Now, this is exciting, man. I appreciate I appreciate you spending time with me. I enjoyed spending time always.