I'm the coach and here's my playbook. We have to do that for companies too. Right? Yep. And our job is we're hired, I guess, as as the coaches from some executive team to train their players on how to use AI. That's right. And whenever people, like, ask, like, oh, what does CADRE do? I think the easiest way to describe it is just we help businesses figure out where, when, and how to use AI. Because there's a lot of great opportunities to use AI in business, and there's also lot of things that businesses want that are not realistic at this point. Companies are starting to think about building knowledge bases or data warehouses because they understand that for AI to do useful things in their business, the AI needs to understand their business. A company like CADRE AI, which is an AI consultancy and implementation firm, where where would you guess we are in the year twenty thirty? I think Hey, guys. Welcome. I'm Keith Jensen, president at CADRE AI. We have I'm Chad Lorelei, founder and chief AI officer at CADRE. Yeah. And we wanted to just chat a little bit today about CADRE and, man, how the company has progressed, where we are today, maybe the best type of people because we are growing extremely fast. I'd love to hear your your background of kinda how you came to it in in the early stages. Yeah. Definitely. We started the company about a year and a half ago and I kinda went public January of twenty twenty four. Not public. We went we announced that we were starting we went to market in January. And so before that, you know, I was, you know, in San Diego trying to figure out what to do with AI. This was maybe a year, year and a half after ChatGPT, and I was like, man, we've got this really great technology. How do I make money with it? You know, I've always been a come from a tech background, software engineer for many years, worked at various startups, started my career at NASA. And I ended up in San Diego. Went to UCSD and was trying to figure out what to do. And I started a community of just folks that were building in the space to try to understand how they were leveraging large language models and how I could leverage them to, you know, make money. And at the time, I I ended up landing a few private equity clients, was growing this community called SDX. And Riley Strickland, who's one of our cofounders, randomly DM'd me on LinkedIn. He was like, hey, man. Would love to chat about the stuff you're working on. And we had a great coffee chat, but I didn't I didn't think anything was gonna come from it. And so I DM'd, I like, hey, man. Great to meet you. By the way, I consult for private equity clients. Do you have any buddies, friends that you can introduce me to? And he said, hey, I know this one guy. He's super well connected, and he's kinda dabbling in the AI space. His name's Grayson. And I was like, I would love to meet him. And so, we all meet, and Grayson just jumps right in. He's like, hey, I wanna start this AI consulting firm, this AI, this tech enabled AI company. And I think this is the right time. He told me his backstory about selling Power Digital, he was like, Riley and I are gonna do this. We can give you a great partner. And obviously, like, I'm summarizing, but this was within the span of like an hour, I was like, what? Like, it's not what I Sounds like Grayson. This sounds like Grayson. Like, I This is not what I was thinking, and he's like, look, come to my house next week, and and let's talk details. And was like, man, don't know if I wanna do this, like, I've got my own thing going. I was talking to my fiance, who's now my wife at the time, and I was just like, looking at Grayson's background, was like, man, this guy's like the real deal. Yeah. You know? And like, we've got Grayson and Riley, who are both proven founders, who can sell, who have built companies, and I really haven't done that, but I know AI pretty well, and I was like, this could be like a good team. It's like a back and forth, back and forth, back and forth, and and finally, Grayson laid out the path. Kinda twisted my arm a little bit, but not really, because I was already all in. He's really good at that. Do that without you knowing it. Yeah. Yeah. And so so he joined forces and and started CADRE, and here we are today with you, Keith, as our president. And so I'd like to hear from you. How did Grayson get you on board? Yeah. So I met Grayson. I was CMO of WAD Capital, so it was a private equity firm. And my job was to oversee at the time, I think it was, I don't know, eighteen or so portfolio companies, companies that they had a majority interest in, and they were trying to grow. My job was to go in and figure out how to scale the marketing efforts at each, get the right marketing leaders, put the right marketing strategy in place, and ultimately, like, track it across the whole portfolio. And one of my one of my colleagues had left WAD Capital and gone to Periscope Equity, and Periscope Equity had just acquired Power Digital. And I think I might get this wrong, but Power Digital, I think, was in, like, the thirty million range at that point. Her name was was Ami, and Ami reached out to me and said, I just met someone that you have to meet. You guys are gonna love each other. And so talked to Grayson, and I think pretty and looking back on it, like, it was totally a sales call for him. You know what I mean? Like, I was just chatting with him as a as somebody. I was like, yo, cool. You're marketing. But but I would say within two months, I had Power Digital in at, like, two to three WAD Capital companies. They were great. You know, really good. After I left WAD Capital, ended up going to Periscope Equity and doing the same thing over there, working in portfolio operations. So got to know Grayson a little bit better. Then, actually, when I moved out to San Diego, rented a house from him. That was fun. And so, yeah, we just got to know each other, and I was at last company I was at was was Engine, a hyper growth tech. They call it a startup, I guess, but they act like a startup, but they're big. Just when I was there, we did a two point one billion series c. But I was with Engine and went and had lunch with Grayson, and he was telling me about what he was potentially building. I was like, you don't need a CMO anytime soon. Like, you know, I'm good. I love where I am. I love the Engine team. I wasn't, you know, looking to leave or anything. And then it was maybe six, seven, eight months later where I was I was ready to to do something different, and I was gonna go do my own thing, go back to consulting for private equity. And, yeah, Grayson brought me in the office, and I don't know, man. Was hard to hard to resist. Honestly, though, the biggest thing, and I've told you the biggest thing that I didn't want was being in the office. I was just gonna breathe out. And now, I love it. Like, it's hard to imagine. I was working remote for twelve years, and now I'm like, what? How did I get anything done? You know? Even just the idea of, like and we said that we're bad we're bad right now about being good to our hybrid That's right. And remote team members. Because, like, we'll get into a meeting, and we'll, like, do our chitchat, and then be like, oh, ****. Start the meeting. Or, like, we'll get off the meeting, and then we'll we'll, like, chitchat more, and we're like, ****. We should've left that person on. We're getting better at it. Yeah. We need to get better because we're already outgrowing our space, and we need to be better about enabling people that are not local in San Diego. But it is really crazy how it went from me going, I don't think I wanna be in the office, to, wow, this is it's I mean, it's hard to be away from home for so many hours of the day. But you like the office. I do. I love Gator. I do, but it's because we get so much done. Right? Like, I really Like, I hate the idea that, like, I'm only home for a few hours to get to spend with my wife and dogs, but, like But, yeah, when you're there, you're just, like, sucked in, and it's, like, you know, it's hard to imagine how you operate it any other way. If you're watching this and and you're remote, just know we're working on it. Yeah. Right. And don't be dissuaded. By the time you see this, we'll have gotten way better at it. I think one of the things that I'd love to to chat about is just like the culture and the the people that have succeeded, people that have done really well because it is a startup, true startup. We've been only around a year and a few months at this point. Did an amazing first year, like, better than we could have imagined. On track to three x in year two Yeah. Revenue. As far as people, I think we're at well, this is you know, we're recording this in February twenty twenty six. We're at maybe forty, forty five people. My guess is by the end of the year, we'd be close to a hundred. We're expanding a new office. So I I'd love to to get your point of view on, like, what's the profile of someone that just, like, crushes it? I think that the word that comes to mind is agency. So they figure it out, you know? Fitso. Figure it the **** out. They they don't need an SOP to go do their job. Right? They use the tools around them, the people around them, to figure it out. They don't complain, frankly, about how many hours they work, how long the days are, because they love it. Right? We're a high growth startup agency. Like, are maybe the three scariest words. Right? Because that means you're working twenty four seven. Right? And so at the end of the day, the people that succeed actually love it. Right? I think you and I are the same way. We talk about this, like, we love work. It's fun. It's a puzzle. There's so many things to solve. It's true. It's true, but I wanna say, like, I I I've been in stages of my career I'm, you know, forty five now. I've been in stages of my career where, like, all I wanted to like, I I just wanted to work constantly. Like, I would wake up in the middle of the night and get up and work. I just love figure I still love figuring out problems. I think that as much as I hate the word life work life balance because I think it's bastardized and people just it's like, oh, well, that means I leave at five. I think I've been pushing as I've gotten older, I've I'm pushing younger people to be like, look, I know I see you at eleven PM on Slack, you know, hitting me up with stuff, and I love that. Like, I get excited about that, but I also just wanna make sure you know that, like, you can ask your leader to unblock you from these things. You don't have to take on everything that comes at you. We're pivoting to this model now, this goal based system of everyone's got these goals, and they know that they have committed. They've signed a paper that says, I'm going to finish this thing by this time, and here's what finished looks like. The thing I keep pushing on, I was just talking to to Trav about it yesterday, was like, if you continue to if you set these goals and you continue to take on everything that people ask you to do, you will miss your own goal and you'll fail. It's on you to set your own boundaries of the day. Right? If you wanna if if if you wanna be done at seven PM or six PM or whatever it is, then it's on you to unblock to ask your leader to unblock you when you have a ton of things coming at you, and you go to them and say, look, I wanna help so and so with this thing because it's gonna help them achieve their goal. But honestly, if I do that, I am not going to be able to hit mine. And it's on your leader to say, you know what? That's more important. Let's push your goal date back a day. Or, you know, no, let me find someone else that can do that thing. They shouldn't be asking you for that. Right? So it's on everyone to kinda have their own, like I said, agency to to determine, like, what their goals are and what they're working on and not burn themselves out. Right? Because we're a startup, we're gonna keep giving you everything. But, like, there's no shortage of **** to do. Yeah. I I think I think the the the point there that that I take away is, like, setting your own boundaries or learning how to do that. And for the young folks, they might not know that. And as leaders, like, it's it's our job to help them set those, and the the goal framework that you've rolled out, I think, is great for that. Right? And also, think something something that was quite interesting for me with goals is, you know, when I try to set goals for my team, I have a tendency to wanna solve all their problems. And I think you kinda do too, right? And so I'm like That's all of let me set up your goals. But instead, right, you kinda like shift in your mind, you're like, no, give them the destination, and have them fill out their journey, and maybe you course correct them a little bit, but don't do their homework for them. And I think that was an enlightening way to to lead. I've fallen in this trap so many times. I was telling the story the other day of one company I was at. I won't name it. But one company I was at, we had this big rollout, big rebrand and, like, big ad campaign that was going out. It was rolling out, you know, Times Square, airports, like, the whole thing. Right? And had a creative team that was gonna work on the campaigns, and my CEO and I had decided we had seen this competitor ad that we're like, ah, this one nailed it. And I don't remember what the four things were, but we like die we like dissected it. We're like, the four things are it does this, it makes you smile, it has this kind of juxtaposition in the image, and this. And we're, guys, here's the destination. It needs to to have these things. And I kept getting, you know, stuff back, and I was like, it's not hitting this one. It's not hitting that one. Like, I just I wasn't getting it from them. And I think I was meeting with them twice a week on it. And I felt the pressure from my boss to get this thing done because the date was coming and we had to roll it out as a big massive roll out. And I just did it. I just I kinda knew what I wanted. I took the picture on my iPhone and wrote the line and built it and, like, showed it to him, and he was like, this is the one. This is what we're going with. And as much as I I think I didn't really care about see, like, when I was younger, was like, oh, I want I want my workout in different places, but, like, I didn't care about this at this point. Right? I was, you know, forty something years old. Like, I just I just wanted to, like, get the job done and, like, please my boss. Right? But what I realized what I did there was I robbed that team of their opportunity to do that. And what I should have done is say, you know what? I'm not getting what I need. Let's meet for fifteen minutes every morning or fifteen minutes twice a day, and pitch me ideas and I'm gonna just guide you, and I'll tell you what's what's right and wrong about it. And I should have leaned in and allowed them to figure it out, but keep, like you said, give them the little redirects of like which way to go. But when you jump in and try to solve it, as much as we like doing that, you're doing two things. One, you are, you know, you're stunting the growth of the person that you're doing it for. And two, you are absolutely causing yourself to do more work and push the things that you would have been doing in that time back. Because it's yes. Sure. You could do it better and faster potentially, but even if that's that's two hours of your work, that's two hours of opportunity cost that you lost of something else you could have done that was of higher value. A big lesson in the last, I guess, year and a half, because I've traditionally been more of an IC and really haven't managed many people or teams, was this shift of, I get a lot of dopamine, and I feel good when I solve problems. And I had to shift my mindset to say, I need to feel good when my team solves problems. And that's a very hard thing for someone that is a go getter and just wants to do it all. And it's not like always for the recognition, although there's an ego part of it. It's just like, I wanna solve it, you know? It's like, I don't want you to solve I want you to beat the game for me. I I don't want you to solve the Sudoku puzzle for me or whatever. Right? Like, wanna do it. Yeah. And and so that was a big big mind mindset shift for me. Yeah. I think the other thing with goals is, like, we can we it's so hard. And I was worried about rolling them out because I was worried it was gonna slow us down. I don't think it has. I think it's actually, like, enabled people to move a little bit faster. But when you paint when you do a goal when we do goals at Cadre, we are we are not describing, like, how do you go do something. It's it's here's what is true. Here are the statements that are true if you meet this goal. And by doing that, you're not giving the solution. You're saying, look. There's somebody in the world that could make these things come true. And they could there's some people that could do it in a week, and there's some people that it's gonna take a month. And there's some people that could do it all by themselves, and there's some people that would need a bunch of resources to do it. Where are you? Right? If someone can do it, why not you? Yeah. And if so, here's what success looks like. Fill out this section that says, what resources do you need? Right? And how long is it gonna take you? Let them set their own goal or their own due date. If it's on I mean, it's on the leader then to determine is that is that the right person to do this? Or can I go get somebody that could just come in and parachute in here for a month and knock this thing out? Yeah. You know? And it's I I I really think it's like it's on us to determine not who are the best people to work here. It's more so what are the goals that we need to accomplish as a business, and who are the best people to accomplish those most important goals? Those are the people that we need to, like, invest in. Hundred percent. Yeah. Very much agree, and I I love the framework. I'm excited to roll it out to the team to see how it affects our, you know, our progress. My hope is that when it catches on, you've got like it's so easy to run to someone's desk and be like, hey, by the way, I need you to do this and this. And we gotta be careful not to be like, everything needs to be a goal. Right? But what I'm hoping happens is that And by the way, for anyone, that that, you know, doesn't work at Kaji, we've been putting them on the wall. Right? So we've been like printing these things out and putting them on the wall. And there's something very analog and cool about like seeing Stuff on the wall now. It's like eighteen goals up there right now. And, like, it's cool to see, like, each person's name and, like, what they're working on. And there's still a lot of people that don't have have goals set. I've had some people come to me, Are we gonna have a hall of fame and, like, a hall of shame, you know, for people that miss and and That'd be pretty funny. You know? I I don't I don't mind that at all. That's one of the things I wanna be like, I really and I I wanna roll out this this whole idea of like, when we miss a goal, like, let's have the person who missed it just talk to the the team and be like, look, here's what I'm gonna do differently next next time. Yeah. Right? Here's what I think I did wrong. I I wanted to impress Chad, so I said it was gonna take two weeks and really, like, I was pretty sure it was gonna take three. Next time, I'm gonna stick to my guns and be like, dude, this is gonna take me three Right? Or whatever it is. Or I I wanted to help so and so, or I got distracted with this. Like, let's let's learn from that. Like, it's okay to fail. I think it's so important. Like, my favorite part of the week is the retros on Friday, to see, like, what's the scoreboard look like? Yeah. Did we we did we hit all our goals? Or in this case, like, you know, did we get all our tickets to complete? And, you know, in a month, it'll be, did we hit all our goals? Right? And the scoreboard thing is is what I love. It's like, you can't play a game. And I really that's kinda why I love work so much. It feels like a game. You know? It's like constantly just shifting, and you're always trying to solve something. But what I love about the scorecard idea is that, like, there's no good game to watch without a scorecard. Right? And we've been trying to move we've been trying to I mean, there's a lot of change management, right, with a startup. You just people get in, they do something a certain way, and then you're trying to be like, hey, we should all standardize the way we do this, and that's change management. And one of the things that's been tough, as you know, is, you know, getting people in the habit of of tracking hours. And we don't track hours because we don't think people are working. We wanna find out how many hours are going towards different, you know, projects. Right? And are we are we, you know, charging the right amount for those projects? But when it comes down to it, it's really hard to get people to take that extra step in that behavior because they don't really they may maybe it's because they don't see the value in it, maybe because it's just hard to, like, you know, make that change. But and it's been maybe a month of us trying to push this, if not a little bit more. We just started rolling out this, like, daily. I'm just kinda posting this thing and being like, alright. Here's everybody's hours that they tracked yesterday. You could see it through the week, and you could see it's red when there's zero. And day two, when we did that, it was like a, like, two hundred percent bump in hours tracked. Oh, what's this saying? It's like, what gets measured gets improved or something. Right? You know, you put it in your face and like, alright, now you've got something that you wanna increase. And all humans, I think, need like a goalpost. I think one question I have is, why do you think people resist? You know, we had a we had a good conversation with our team, you know, a month ago or something. Yeah. And you made this analogy of like, hey, like, I'm the coach, or I've been brought in to be the coach, and here's my playbook, and I don't know why you guys aren't doing certain things. But you had like a few reasons, right? I think they were good. So I remember man, this feels like a year ago. Yeah. That was probably a month ago. I remember having that conversation, and my frustration was, like, I'm you've got Grayson who's, like his job is he wants to build this company. He has a vision for how the company is going to like, he know he knows the exit he wants. He knows what he wants to build. He hires his people, his he's if he's the general manager, he's high hiring the coach, he's hiring this. And that coach is supposed to then figure out, like, what's my overarching goal, and what do I need to do to put in place in order to to get there, and then set those goals and hire the right people to to do them. And it was frustrating when people were resisting things. They weren't tracking hours, I said, please track hours. They weren't doing this. And I don't remember all the reasons that I gave, but like, top of my head, I feel like it's it's like it's either that they don't believe these are the right plays, so they're gonna resist. And the analogy of that is a coach saying, hey. Let's go run this play as a, I don't know, football game, whatever. They let's go run this play, and the team just does something different. Better analogy would be that whole team doesn't do the same thing different. They don't run a different play. They all just do something. I mean, can you imagine if there was just coach calls a play, then everybody out there just does whatever the hell they want. Like, it's chaos. Right? And so it's either that they don't believe in the coach and their their ideas. They don't know how to change their own behavior. Could be part of it. And there's a lot of different reasons, but I I ultimately said, like, look. We are either gonna get to a point where I'm gonna go, guys, this has been months of me calling plays, and you guys aren't running them. You need a different coach. Like, I shouldn't be the coach. Because if e whether it's what no matter what the reason is, if I can't influence this change, someone else needs to come do it for the good of the company. You know, I always think about, like, roles are are temporary. You know? I'm I'm the president of Quadra AI right now. I may not be. I likely will not be. When we get to a much bigger point, there's probably gonna be someone way more qualified than me to have this role, and I will happily step aside, to to let someone better do it because it's for the good of the company. But, yeah, I think I think when it comes down to it, like, you have to you have to to influence that change, and if you're not the right person to influence that change, then you gotta get someone in that that's going to. Yeah. What's interesting is, like, we have to do that for companies too. Right? Yeah. And our job is we're hired, I guess, as as the coaches from some executive team that to train their, you know, players on how to use AI. And it's tough, because we get kind of the same thing. They don't understand it. They have a rigid mindset. Maybe they wanna they're resisting it because they're scared, maybe they don't believe in it. Right? So it's interesting, because like a lot of things we face is a microcosm of the things that we do at companies. That's right. And whenever people, like, ask, like, oh, what does CADRE do? I think the easiest way to describe it is just we help businesses figure out where, when, and how to use AI. Because there's a lot of great opportunities to use AI in business, and there's also a lot of things that businesses want that are not realistic at this point. Like, everyone's talking about just agents as employees just going and be you know, having agency to do whatever, and and we haven't seen that really come to fruition yet. The other big thing that every client seems to want is just a a massive database. Well, actually, they don't even say the database part. They say, I wanna be able to talk to all my data and just get correct answers, which we know, like, you can't just take a whole bunch of MCPs and connect various things in and be like, hey, LLM, answer anything I want. Right? You're gonna have to build this data infrastructure with Snowflake or something else in order to rag model to me in order to order to, like, actually get that correct information. And with AI, as soon as you get a wrong answer, you're not very forgiving. Yeah. It's like, well, this thing sucks, you know, and you move on. So you're right. It's our job to not only help businesses figure out how they can use AI, but then we go and build it for them. In some cases, we'll find a product that's on the market and implement it. In other cases, yeah, we'll just go build it. And, that's another great topic of, like, how fast Yeah. AI has come along just to allow us to, you know, speed up the way that we can work, and the way that we can enable these really cool micro SaaS applications for our clients. Totally. Yeah. I don't wanna give all the secrets out, right? Because then people are just gonna, you know, build a Cautery two point o. No. I think, look, you know, the the the pace at which software engineering is changing is is mind blowing, and as someone that's written software for, I don't know, eight or nine years of my life, to not have written a line of code for, I wanna say, like, a year, and instead, I'm I'm just brain dumping my thoughts into AI, and it's creating things with Clawd of Code and some of these great tools. I just didn't believe we'd be here this fast. What it means for our business is, you know, we can hop on a call with a client. They can describe what they want, whether it's an AI sales coach or an AI content marketing engine, and pretty quickly, we can turn that into a product, or at least a UI that can validate, is this the right thing? And then very soon after, into a product with a full back end that plugs into another AI to process, you know, transcripts, or process content, or do whatever they want. That's the tricky part. Right? It's like, you've you've we've all seen this where you ask AI to build something for you, and then you're like, oh my god, this is great. Do it again with this other use case. Yeah. And it's a little different. The icons are different. The colors are different. The the the way it's like the the way it's gonna store, the schema, the the database, it's all different. Having this I think you've called it like a framework, or like having this framework behind the scenes of like, these these are your constraints. Like, you have every ability to like code whatever you want, but these are the constraints of like you need to fit in. I think that's, like, one of the the critical parts that, when businesses try to adopt this themselves, they're gonna find very quickly that, like, you can yeah. Sure. You if you have a product already, you can have Cloud Code go build an additional product. But to have it integrate with your product, there's a whole framework that needs to be built underneath in order for it to understand what it should and shouldn't pull from Yeah. And, you know, how to build on something that already exists. Yeah. I think the closest analog to this is, you know, companies are starting to think about building knowledge bases or data warehouses because they understand that for AI to do useful things in their business, the AI needs to understand their business. So maybe it needs to understand their financial models in Excel. Maybe it needs to understand how they respond to customer service tickets. Maybe it needs to understand their sales philosophy. Well, code is no different. Right? For an AI to generate code, it needs to understand, you know, the code knowledge base. How do we write code in our company? What is our tech stack in our company? What are patterns that we like? What is our design kit? You know? And so we're just building that layer at Codre, which is our code knowledge base for Cloud Code to generate templatized, standardized code. Looks the same, feels the same. And when new developers come, they learn our way, and they can quickly build, and everything looks good. It's nice. It passes its tests. Well, the the benefit here is having the constraints because That's right. We've tried, like, when we when we start I mean, you guys started the company. When I came in, I remember actually, I remember talking to Grayson when he was thinking about it. I was like, my god. The it's like endless. The ideas of like, you can just have AI do the whole business. Like, if we like, for marketing, like, why wouldn't AI be listening to, you know, call recordings from sales and immediately going and building campaigns, figuring out what the common objections might be, and then building camp whole campaigns around, you know, outbound email, blogs, whatever, just to handle those objections. Now we could also feed it in to say, build an objection handling tool that, like, goes on the website. But there are so many things that you can do and can automate. Obviously, picking the most important ones are are critical because we don't have an endless amount of resources to put into that even though it is our own business. We have to serve our clients. I think the the we've learned our own lessons of, like, don't let AI write contracts. Right? Don't let AI do this or do that. And so that's also, I I think, where businesses are struggling, where they just don't know where not to use it. And we've we've, again, faced our own challenges with that. Yeah. I mean, I I think using AI in a domain that you're not super well versed in for example, like, you know, the the contract example. I think that we didn't really have a standard contract template, and we didn't really understand what to put in it at times. And so all of a sudden, you're just delegating to an AI that is just gonna take the average of all the contracts in the world and say, here's what I think it should be. And if you don't understand the domain, and you haven't written one yourself, and you just say, go write it, guess what? It's it's, you know, it's it's as if you just pulled it from from some random company and said, hey, here's our new contract. Right? Grayson always has a good analogy of he says, like, with interns. Right? It's it's it's more work to have an intern when you when you're not giving them a very specific thing to do. Interns are are phenomenal when you come in and say, here's the one thing you're gonna do over and over and over until you're great at it, and then I'm gonna give you something I remember early days when I would have interns, and I've stopped for a very long time because of this. But I would have interns, I would be driving in, and I'd be like, ****. What am I gonna have them work on today? You know? And, like and then I realized, like, I get in, and I'm like, alright. Here's what I want you to do. And I start giving them all this context that they need in order to do this thing, and then it comes back at the end of the day, I'm like, this isn't at all what I was hoping for. I could have just done this myself. And it's like, well, yeah, dummy. Like, of course, you you should have just done that yourself with the time it took to explain it if that was a one off thing. But if that was a thing that they should do over and over and over, take the time, sit down, do it with them once. Make sure they'd see how you can do it, and then watch them do it, and then let them do it on their own. And I think AI's similar in a lot of ways. If you let it just make decisions that you haven't decided for yourself, you're gonna have a worse result. I think that's right. Yeah. I I almost see Cadre as, like, a talented a HR company for agents that we deploy. Right? Like, we need to train them. That's right. We need to to make sure that they're they're healthy and happy. Right? That they have enough tokens to eat kind of thing. But I think you're right, like, especially in new domains, you have to do it once or twice so that you can validate the output. Right? And so I think where we're headed with agents is the cost to generate is is going close to zero. Like, we can generate anything all the time in parallel at scale, But the bottleneck now becomes reviewing. Right. And because we get so used to just generating, we get lazy on the review side. And so we're just gonna trust the generation, and that's when things kinda go sideways, you know? When I get something given to me by even one of our team members, right, that's like, hey, I built this thing, or hey, I I I mapped out this this workflow that you had mentioned. And I look at it, and then I I see something in there. One thing in there that I'm like, what's this? They're like, oh, AI put that in. I I don't know. Don't worry about that. I immediately go, well, not trusting any of this. Right. You didn't review it. Yeah. Right? Like, that's the thing is that, like, AI can get you. And I think this also is something that a lot of businesses may not realize when they're like, oh my my partner's little brother, Ken, has been doing this thing. He built this thing. He did that. Yeah. AI can get you to, if you think of, zero to a hundred, AI can get you to ninety percent so fast. That last ten to whatever however percent you wanna put, that is the hardest, and that takes the longest. That's right. Yeah. Even like we we talked some of the things that you it feels like it'd be so simple. Right? You can create a custom GPT and load an HR handbook into it and say, hey. I now have an HR chatbot. Go ahead and and, you know, give this out to the team and and let it, let people chat with it so our HR team doesn't have to do it. They say thirty percent of HR team members' time is answering questions that are answered in the handbook. But if you write a very simple prompt, hey. Here's my handbook. You know? Answer questions based on it. You're you're gonna get wrong answers probably fifty percent of the time or more. Yeah. Right? Yeah. That prompt that we ultimately ended up coming up with to make that HR handbook work, it was in JSON format, and it was, like, eight thousand rows long. Right? Because it it has to keep like, you have to tell it. Read the whole thing before you answer. You know, there's instances where it might go, hey. Do we have Columbus Day off or something? Or and then it it it reason is it's something see, something at the beginning of, and as part of your package, you have major holidays off. And it just says, yes. You do. Yeah. It doesn't reference the table on page eighty seven that shows that you have to work on Columbus Day. Right? So it's those types, the prompt the prompting of how much or how much detail you have to put in the prompt in order to get this to be accurate at the level that it's valuable to you is insane. Well, I think, like, AI, just like humans, are in is inherently lazy. It will try to get an answer as fast as possible. Just like an intern. Right? Unless you explicitly tell them. So really, a prompt is an SOP for an agent. And the more time you invest in that, the better the result will be. Just like an intern. If you give them the step by step process, and you say, go do this. Now, if they don't do that, right. Maybe maybe this is a different conversation. But if they do that, you'll typically get a good result. And so I see our job as, again, like, we are training, teaching, building SOPs for agents, and using, you know, companies' tribal knowledge to influence the agents to work well within this new environment. With you having such a good pulse on kinda where AI is and and isn't, where would you guess a company like Cadre AI, which is an AI consultancy and implementation firm, where where would you guess we are in the year twenty thirty? Like, what are we doing? What are we selling? How are we enabling businesses better to use AI? I think I'll preface this by saying we are in a very good spot because we can shift and mold to changes in technology and in AI. So we're built for. If it that's what we're built for. We're we're resilient by nature. Right? And when you think of these new startups that are building products, where now you can write code very quickly, and you might be one model update away from an incumbent just eating your lunch, which has happened time and time again, if you just look at the last YC batches, we don't have that problem. Right? And I was talking to one of the partners at YC, and he said, look, if you froze model development today, it would take roughly ten years for large businesses to catch up. What that really means for CADRE is we've got a ten year arbitrage window. And if we can't sell and become a unicorn in the next ten years, then we're just failures. Unicorn? Oh, it's gonna be way sooner than that. You crazy? Yeah. I'm very optimistic. I think by twenty thirty, we will probably be doing agents will probably be doing most of the work that we do today at Cadre. Our jobs will be to hop on calls or, you know, do in person trainings, teach people how to use AI. But the products or the services or the tech that we build will probably be happening in the background. And we'll be having more of these conversations with humans Yeah. And making sure that they feel comfortable with AI, and and figuring out what to do when AI takes a lot of your job. You know what I find interesting is, like, I've always been resistant to giving, like, context about my life to a computer. Right? Even, like, cookies and that type of thing. But I but with how valuable AI has been, LLM, just being able to, like, chat with it on the way to work. Know, a lot of times, like, I have a I have one that just says solutions notes, I just didn't know who all everybody in the team is, and I'm just, you know, riffing with it of, you know, stuff. And but it's like you have to give it that context Yeah. Every time in order for it to be helpful. Like, the fact that it's so helpful makes me feel like if if someone came up with a wearable or something that I just wore all day at work and it just ingested everything. It would have that context. It would be able to show me better where my faults are, you know, where I could have communicated better, action items that I didn't jot down in my notebook. Right? There's just so much it's crazy how much my mindset has shifted on my willingness to open it up to my life just being recorded Yeah. You know, because of the value you get from it. I I'm the same way, I think. I'm I'm way more open now. And honestly, I don't have many skeletons in my closet anyway, so I'm like, you know, open book. Yeah. You know, if you wanna steal my data, steal it. As long as I can be more productive and, you know, create more value, then that's fine. Tricky part comes when advertising and everything taps into all of this, and now you start realizing like how curated everything is to you. Yeah. You know? Like, you know, there's the I'm forgetting what the the biased name is, but it's like, you know, you, with social media and all that, you're getting fed the stuff that already it knows you want. Right? It's not typically sending you things that are contrarian to your beliefs. Right? It's like it's like reinforcing what you already believe. Imagine if now your whole world, every ad you saw I mean, remember years ago, probably ten years ago, they were, you know, talking about beacons and how, you know, you'd be walking. You should we we're at a point soon where you're gonna walk by a department store or somewhere in Times Square, and the whole LED is going to be just for you because it's gonna know, you know, who you are, and it's gonna gonna serve that ad to you. And it's like, we're we have not gotten anywhere close to that. Not a good idea now. Right? It's not a bad I know now we can probably do that with how quickly we're able to generate images and all of that. But, yeah, I think that's gonna be the the downside of of that, you know, giving the data, being willing to give that give that amount of data. But, also, you think about, like, robots. Like, Elon Musk is shutting down the the Tesla Model x and s just to get the warehouse space to build robots. And so if we are to believe that robots are gonna be out there, then, like, literally everything is gonna be recorded. Right? They they have they have to be aware of this their surroundings. Like, in China's already there. Right? You know? And they're they're much more modernized than us, so they've kind of accepted, and and they enjoy the productivity, and they're just like, you know, it is what it is. So Yeah. Alright. That was a fun chat. Yeah, man. Yeah. Alright. Cool.