// life timeline
Watches a TI-99 load games off a tape recorder and starts saving his money
Buys his own TI-99 and teaches himself BASIC at 14
Navy electronics technician on the Snap 2 mainframe, dBase III on the side
2025February 16: pitches an idea to an AI from his phone and is renting servers by the weekend
2025Discovers Claude Code
2025August: starts building Mandrel
2025October: Mandrel is done, and everything he builds goes through it
2025Builds Squire, his personal assistant
2026Backstory, OS memory for Omarchy
2026Opens up Mandrel as mandrel-oss under MIT
// 00
The rundown
Brian Jones is a rock star. He taught himself BASIC on a TI-99, ran reel to reel backups on a Navy mainframe, and spent his days selling flooring across a big territory. Then on February 16, 2025 he typed one idea into an AI and never stopped. He has built every single night since. The result is Mandrel, a memory system for AI agents that he designed, shipped and lives inside every day. Ask him why he went so deep and the answer fits on a sticker:
"I didn't want to build a toy."
“Because I didn't want to build a toy.”Brian Jones · L200
// 01
A TI-99 and a tape recorder
Brian spent part of his childhood in the back of an amusement company, cleaning pinball machines and collecting free games as payment. In the '80s, he says, that was the coolest thing ever. A guy who worked back there had bought a TI-99 and was loading games off a tape recorder, through a homemade cable built from Radio Shack parts.
That was it. Brian started saving his money that same day. He bought his own TI-99 at 14 and taught himself BASIC, building little games. Then the Commodore 64s showed up in high school, along with Hello World.
// 02
Reel to reel on the Snap 2
Brian joined the Navy and went to school for electronics. He was an ET. One of his jobs was the Snap 2 mainframe, the supply computer, where he ran the backups every night on reel to reel tape. Along the way he wrote dBase III code, and he was there when the internet was just getting going on AOL dial up. "We've seen a lot of change, haven't we, man?"
Deep down, he says, he is a troubleshooter and a problem solver. He used to take stuff apart all the time. That instinct never left.
// 03
February 16, 2025
Brian remembers the exact date. He had an idea for his day job, all the paperwork and antiquated stuff every sales rep hates, and he started banging it out with an AI on his phone. That weekend he was renting servers.
The first wall he hit was memory. Every new session started from zero, and with a full time job he only had four or five hours a night to build. "So I got obsessed with, you know, how do you solve this problem early on, right?" Then he found Claude Code. "That was kind of it for me."
// 04
Not a toy
He did not rush it. Brian spent months learning before he wrote the tool he uses every day. He wanted real memory and real project management, a system that knows not just what got built but why. "Yeah, you can build something, but how do you maintain it, right, after?"
He started Mandrel in August 2025. By October it was done. "And then I just dog fooded the heck out of it, man." Today, he says, everything he builds goes through it. Mandrel runs on Postgres. It is not just markdown files. It uses salience factors so old memories decay, and every night it goes to sleep and wakes up with a living summary of what it knows. He can text search his whole history or pull any decision by ID. The open source version, mandrel-oss, gives AI agents persistent memory and a decision graph over MCP, under the MIT license.
Mandrel let him build more. Squire, his daily driver personal assistant, runs on a memory design that is all his own. He runs it all from tmux and steers long running sessions from his phone with remote control.
His big rule is to make things as deterministic as possible. Gates, tests, CI flow and hooks, never just prompts, because prompts get pushed down in the context and stop being prominent. "You don't need it. It's cheaper, number one, and you get better results."
On where it all goes, Brian thinks privacy is a huge concern and local will win once somebody puts the right package together. He also sees a gap most businesses are missing: build a back end for agents right now and be the first one there.
// 05
Sunday Builders Club
Brian is an Omarchy builder and a regular in the Sunday Builders Club, where the group was talking local versus frontier models the night before his interview. "That's why I love that group. It's a pretty good group in there."
He has built several things for Omarchy, including Backstory, OS memory for Omarchy that records the why, with outcomes, under every agent. He has a merged PR on a community Omarchy plugin too. And he keeps learning, every day, with a session open right now going deeper on databases. "I think you got the ability to pretty much learn whatever you want to learn right now, you know?"
He wants a foot in both worlds. A drive on the parkway, then back to the terminal. "I want nature and then I want tech."
You are a rock star, Brian.
- X: @RidgetopAI
- GitHub: RidgetopAi
- Site: ridgetopai.net
// 06
In his words
"Because I didn't want to build a toy."
"But everything I built goes through it. Everything."
"That's why I think it's important that you spend time on your gates and your tests and, you know, your CI flow."
"I think you got the ability to pretty much learn whatever you want to learn right now, you know?"
// their setup
Setup
hardware
tools & models
mandrelsquirebackstoryclaude-codetmuxpostgresmcpremote-controlclaudetypescriptgoagent workflow
Building with AI every night since February 16→2025
// full transcript (cleaned)
Show the cleaned transcript (272 lines) · open full page
L1[redacted: private, lines 1 to 42]
L43Brian: That's kind of, you know, where I think we're all trying to head to, for sure. It's just, man, it's time for me. Because I work a full-time job, and this, you know, this is not my field, right?
L44Brian: Like, I have some, I have a history with some things like this, with tech and, you know, and self-taught and that kind of thing. But, you know, I'm a sales rep for a flooring distributor.
L45Host: Nice.
L46[redacted: private, lines 46 to 50]
L51Brian: And so, yeah, I think we're all trying to get there though, man. Some of us faster than others.
L52[redacted: private, lines 52 to 57]
L58Brian: But I've learned a lot. I've been doing this for since February 2025, every night.
L59Host: Right. Are you able to wrap in your sales business into it somehow?
L60Brian: Well, that's kind of how it all started. It's, you know, I saw an opportunity. I thought, you know, saw an opportunity, and I started down that path.
L61Brian: But the first thing that I, you know, discovered was, because February 2025, and it was [agent] of all, you know. And I had discussed an idea with [agent] because at the time, that's really all I knew.
L62Brian: But anyway, back then, you know, there was no memory. There was no, every time I started an idea, I started to talk about something, and I thought, okay, then I had to do it all over again.
L63Brian: You know, and then being part-time, you know, I had to come in and only work 4 or 5 hours a night.
L64Host: Right.
L65Brian: It just was a lot. So I got obsessed with, you know, how do you solve this problem early on, right? And then, of course, I discovered Claude Code, and, you know, that was kind of it for me.
L66Host: Yeah.
L67Brian: Once I got online.
L68[redacted: private, lines 68 to 80]
L81Brian: Listen, I get it, man. Because this stuff, first want to say, if I had to sit here and code by hand.
L82Host: Right.
L83Brian: I think I could do it. I could, I mean, I could do some things, I think, if I spend enough time. And what I'm saying is, I understand the effort.
L84Brian: I self-taught myself basic.
L85Brian: So I was 14 years old.
L86[redacted: private, lines 86 to 88]
L89Brian: Yeah. Yeah, so I bought a TI-99, you know, when I was 14 years old. And I self-taught myself basic, right?
L90Brian: Building little games and stuff.
L91Host: Yep.
L92[redacted: style, lines 92 to 93]
L94Brian: Yeah, no, but I think I was trying to say that, oh, I know my point.
L95Host: Can we turn this into an interview for Omarchy Builders?
L96Brian: Sure, man, if you want. Yeah.
L97Host: Do you know what Omarchy Builders is?
L98Brian: Uh.
L99Host: How long do you have? Do you have another 30 minutes or better?
L100Brian: Yeah. Yeah, yeah.
L101Host: Okay. So get on your browser and look at Omarchy. Do you use Omarchy?
L102Brian: Yeah. Oh, yeah, yeah. I'm in our, I'm in the Sunday Builders Club.
L103Brian: Is that what you,
L104[redacted: private, lines 104 to 111]
L112Host: Go look. Pull up a browser and open up omarchy.builders. It's a strange URL, but it works.
L113Brian: Builders. Okay.
L114Host: So you would be, I think it's number.
L115Brian: Oh, yeah, no, I did look at this. Yeah, yeah, yeah.
L116Host: Yeah, so like, can I feature you on this? It would be amazing.
L117Brian: Sure, man. That'd be great. Yeah, I appreciate it.
L118Host: All right, well, we've already,
L119Brian: Like, this, I'm not going to lie, it's really pretty good timing because I just finished what I use, my main memory system. Like I said, I got obsessed with it, but I just finished my open source version, and I just launched, like, I just made it public last night. So it'd be good timing, you know?
L120Host: Well, that's awesome. We can promote that as well.
L121Brian: No, I don't, I'm not saying I want to do that. I'm just saying, you know, we talk about, you know, I got it in my mind to talk about anyway.
L122Host: Well, then let's do it, man.
L123Brian: Let's talk about whatever, man. Yeah.
L124[redacted: private, lines 124 to 125]
L126Host: Cool. Well, what was the first computer you ever saw? And what, like, what, and there's a theme that I've heard in everybody that I talk to is that the first computer they saw was not the first computer they got to play with.
L127Brian: Ooh, first computer I've seen.
L128[redacted: private]
L129Brian: I'm going to say it was a TI-99 because I knew an amusement company. They had video games, right?
L130Host: Right.
L131Brian: They, pool tables, video games, pinballs. And so there was a, and you know, when I was a kid, you know, I was young, and I used to go over there, and I'd do a little work in the back, you know, cleaning pinball machines, and of course, getting to play free games, right?
L132Brian: Which in the '80s, that was the coolest thing ever, right? So, yeah, no, we, there was a guy that worked in the back that I kind of, I don't know, we just connected to him, I guess, a little bit.
L133Brian: He was nice to me, right? Like, he was nice to me as a kid.
L134Brian: But he had bought a TI-99, and he was, you know, he was loading games with a tape recorder, right? And a homemade cable that you had to go to Radio Shack and buy the parts for and make yourself, right?
L135Brian: Did all that, right? So that's how I think, I think that's the first one I saw.
L136Brian: And right then, I was like, started saving my money, right? I got to go buy TI-99 because this is cool.
L137Brian: I seen this guy making this game, right? And I'm, yeah, that's, and then, you know, I was right at that time, too, is when computers came in high school and Commodore 64s and, you know, all that stuff.
L138Brian: Hello World and all that, right?
L139Host: And then, so then you're not a coder. So let's jump forward to when the first AI that you learned about and that you started using, maybe like ChatGPT-2 or 3, like the early days.
L140Brian: Yeah, no, again, I started this, I can tell you the date, February 16, 2025, right?
L141Host: No kidding. That's amazing.
L142Brian: Well, I know the time because it all came about because of an idea I had for my work, what I do for a day job. And the only AI I knew then was [agent]. So I jumped on there and started banging out, you know, it was actually on my phone, started banging out my idea and, hey, what about this?
L143Brian: What about that? And, you know, you know how AI is, man.
L144Brian: It'll, especially then, you know.
L145Host: It's like a fancy.
L146Brian: Yeah, it was just off the charts. You could do anything, right? Yeah, we can do that.
L147Host: You're amazing. That's the most incredible thing I've ever heard.
L148Brian: Like, I think that that was on a Tuesday or a Wednesday. That was February 16.
L149Brian: I'm pretty sure it was a Tuesday. And that weekend, I'm renting servers.
L150Brian: [agent] had me renting servers, right?
L151Host: That's awesome.
L152Brian: Like, I said, I dabbled. Like, I wrote some dBase III stuff way long time ago for when I used to work at, I was in the Navy, and I went to study electronics. I was an ET.
L153[redacted: private]
L154Brian: Yeah, so one of my jobs was, well, yeah, one of my jobs was we had a supply computer that we used called Snap. It was actually the Snap 2 mainframe, right? So I kind of went to school for that, ran backups every night, reel to reel backups.
L155[redacted: private]
L156Brian: And on the side there, I kind of dabbled. I had a cool boss, and he liked computers and, you know, and all that stuff in too.
L157Brian: And, you know, the internet was just getting going, and, you know, you still had AOL dial up and all that stuff, right? We've seen a lot of change, haven't we, man?
L158Host: Yeah, we have.
L159Brian: Over the years.
L160[redacted: private, lines 160 to 162]
L163Brian: Yeah. I'm hoping we can hang in there long enough to see. It's exciting right now, man.
L164Brian: It is exciting.
L165[redacted: private, lines 165 to 175]
L176Host: And what did it become?
L177Brian: So, yeah, I took this idea, which was, you know, like I said, I'm a sales rep, so I was trying to.
L178Host: I can tell.
L179Brian: I was, yeah, I was thinking about dealers, you know, and I'm thinking about my dealers. And when you're a sales rep and you have a territory, it's kind of like you're running.
L180Host: I get it.
L181Brian: Running your territory.
L182Host: You still got two spiffs and discounts and.
L183Brian: Yeah, yeah. So I'm like.
L184Host: Cost of goods, all of it.
L185Brian: There's so many things I could do here, right? We could, you know, paperwork. We all hate paperwork as salesmen, right?
L186Brian: And antiquated stuff. So anyway, that's kind of how it went.
L187Brian: And again, February 2025, there was no memory. There was no even, you know, markdowns on the frontier side.
L188Brian: Like, they weren't using anything. You were starting zero every time.
L189Host: Every time.
L190Brian: Every time. So that was the first problem, right? And I think deep down, troubleshooter and problem solver, right?
L191Brian: Like, that's what it's really in me, right? I used to take stuff apart all the time.
L192Brian: So I saw that problem. Like, I can't build this without a memory or a way to manage, even after I build it, right?
L193Brian: Because I learned early on, you know, it's not just, can you build something? Yeah, you can build something, but how do you maintain it, right, after?
L194Brian: How do you make it better and not lose your place? And, you know, how do you, why did you do something?
L195Brian: You know, why did you do that back then? You know, all of that stuff.
L196Host: Stay scarce and success is recorded.
L197Brian: Yeah, so all of that. That's what I started working on. But I spent a long time before, so I said, February 2025, I didn't start building this tool that I use daily now until August.
L198Brian: I really wanted to, like, I wanted to, I wanted to build a toy. I didn't want to build it.
L199Brian: No, no, no. This a year ago.
L200Brian: Yeah, 2025, August. Because I didn't want to build a toy.
L201Brian: Like, that's not what I wanted. I wanted to build something that is going to be as close to a memory system that I needed.
L202Brian: And project management too, right? It was project management as much as memory.
L203Brian: And so August, I started building it. I think October it was done.
L204Brian: And then I just dog fooded the heck out of it, man. I mean, it's just been nothing but dog fooding for it then, you know, ever since.
L205Brian: But everything I built goes through it. Everything.
L206[redacted: private, lines 206 to 212]
L213Brian: So I just got obsessed with memory. I got obsessed with making this product right. And then because it's allowed me to build some other, I built a personal assistant.
L214Brian: You got [host]. I built something called Squire that, you know, was kind of my daily driver, you know, personal assistant.
L215Brian: Now I've kind of, it needs a lot more work to it. It's not as refined right now.
L216Brian: But the memory I use for it was my own design as far as it's not just straight markdown stuff. We're putting salience factors in so you get some decay and it goes to sleep, you know, what we call sleep at night and wakes up with this living summary every time it starts.
L217Brian: So it kind of has an outline of how, you know, that it knows.
L218Host: Are you using SQL, LIDAR, Vector, or any of those?
L219Brian: Postgres.
L220Host: Okay. Wow, you went big.
L221Brian: Yeah.
L222Host: That's big. You start off from the beginning.
L223Brian: Well, that's been, that's what I said. I did a lot of learning, a lot of learning first, a lot of before I even started because.
L224Host: You know that that will scale to 100,000 users.
L225Brian: Oh, yeah. If you do it, if you do it right, if it's done right. Yeah.
L226Brian: So, but yeah, that's, you know, and then I've just, I built several things for Omarchy now.
L227[redacted: private, lines 227 to 228]
L229Brian: You know, and see all that stuff that I build and I've done, like, it's all recorded in Mandrel. I can go there right now and we could, you know, text search it. I could search it by ID.
L230Host: Right.
L231Brian: So, yeah.
L232[redacted: private, lines 232 to 254]
L255Host: What was your X handle?
L256Brian: It's @RidgetopAI.
L257[redacted: style]
L258Brian: Yeah. It's hard, I know, man. It's hard to put a, you know, you got somebody with a handle over here and a name.
L259Host: You're RidgetopAI.
L260Brian: Yeah, that's me.
L261Host: Okay. I disagree. You're doing amazing s***.
L262Host: Okay. You said it earlier before, and it is so cool, and take this the right way.
L263Host: The faces do not match anything on X.
L264Brian: No.
L265[redacted: private]
L266Host: And like, you impressed me, man.
L267Brian: Well, I appreciate that. Thank you. It's a lot, man.
L268Host: I can't believe you can say that. I can't process that you're Ridgetop.
L269Brian: It's a lot. It's a lot, man. And it's, I learn every day.
L270Brian: Like, every day. And that's one of the exciting things for me is like, I'm learning constantly.
L271Host: Yeah.
L272Brian: You know, and with AI, you know, I've got a session going now. Just, I want to go back over databases. Like, I want to know databases deeper, right?
L273Brian: I want to understand.
L274Host: Right.
L275Brian: So it's, that's the thing. That's the stuff that excites me about this time, especially for young people, right? Because you have this tool now sitting in front of you that, you know, if I had this when I was 14 years old, now I have a great life, and I'm not saying that, but I'm just saying my path would have been different for sure.
L276Brian: But it's just, I think you got the ability to pretty much learn whatever you want to learn right now, you know?
L277Host: Right.
L278Brian: Your interest. You can, what am I trying to say? You can enhance your individualism, right?
L279Brian: Like, whatever it is that you have the most interest in, you can just zoom in right there.
L280Host: That's crazy. That's crazy. So what do you see?
L281[redacted: private, lines 281 to 294]
L295Brian: I'm a tmux. I'm a big tmux guy.
L296[redacted: private, lines 296 to 300]
L301Brian: Yeah. No, I do the same thing. I can do the same thing with tmux.
L302[redacted: private, lines 302 to 305]
L306Brian: No, absolutely. Absolutely. And that's why, you know, not to keep pushing on the memory thing, but I focus there so much because I don't think it's totally solved yet.
L307Brian: But a lot of everything else is going to be solved. And it will be.
L308Brian: It will be shortly. But I agree with what you're saying.
L309Brian: The way we're going is you're not going to see the internals as much. And listen, I don't as much anymore, either the way I work.
L310Brian: I mean, I have my whole runtime around, and, you know, I use Slash remote control, and I use it from my phone daily, you know, and I can keep them moving. I can keep them going a pretty long time right now, especially with, you know, how we work.
L311Brian: So you can't read it all. You can't read it all, right?
L312Brian: And you don't. And if anybody says they are, then I find it hard to believe.
L313Host: Like, whenever the 1200 baud modems came out, you no longer could read it all.
L314Brian: That's why I think it's important that you spend time on your gates and your tests and, you know, your CI flow. You know, if you have, I try to make things as deterministic as I possibly can, right? My runtime set up that way.
L315Brian: Mandrel is set up that way. Hooks, you know, don't leave it to pros, right?
L316Brian: That's one of my big rules. Don't, because they're going to eventually do what they want to do.
L317Brian: And that's not the right way to say it. They forget, right?
L318Brian: It gets pushed down. It gets pushed down in the context, and it's just not as important anymore, so it's not prominent.
L319Brian: So deterministic things, I've sub-quantic the SNAP model that they came out with. I instituted that in Mandrel to make some decisions that we could take out of the reasoning.
L320Brian: You don't need it. It's cheaper, number one, and you get better results.
L321Host: Right. Now, there's, when me and you walk out the door, every person you meet in your daily life, you know more about AI than they do, just by where you live.
L322Brian: In my territory, I'm not a braggadocious person, but in my territory, where I live.
L323[redacted: private, lines 323 to 327]
L328Brian: No, not yet, but I am starting to have that conversation. Again, I'm pretty conservative with, like, I want to make sure I'm comfortable that what I'm going to talk about, I know, right? Or I know what best I can.
L329Host: We'll cut your boy in us.
L330Brian: Yeah. I just don't want to, I'm a salesman, but I'm not that kind of salesman.
L331Host: Right. The honest guy. Like, honest type that does what you say.
L332Brian: I'm more of your, I'm your liaison. I'm your customer service guy. I'm a white gloves, right?
L333Brian: What you need, I'm going to be there. You need a price?
L334Brian: Here it is. You know?
L335Brian: So I've been having conversations with my dealers now, starting to just say, hey, you guys using AI or, you know, what do you think about AI for your business? What do you, you know?
L336Brian: So I'm starting to, because mainly not because it'd be a conflict and I wouldn't do it, but I'm just interested in who is starting to use it, right? And who's not?
L337Brian: And my dealers are all over the place.
L338Host: The industry they're in is not acclimated to technology in the first place. You got a big wall.
L339Brian: Yeah. It's a lot of, you know, my dealers are mom-and-pop stores.
L340[redacted: private]
L341Brian: Yeah. Absolutely. That's right.
L342Brian: And the manufacturers I represent only sell to them, too. They don't sell to Lowe's and Home Depot and those kind of people.
L343Brian: So I'm pretty proud of that. But yeah, they're starting to, though.
L344Brian: There are a couple. There are a few.
L345Brian: There are some younger, but, you know, it's an age thing, too.
L346[redacted: private, lines 346 to 352]
L353Brian: Yeah. Which, listen, no, I was going to say, it's not, I was going to say, it's not a bad thing because you're going to have, you're going to have a generation that holds on to some culture, right? Some of that old culture.
L354Brian: And you need that. You got to have that, you know?
L355Brian: What is it? The 70s, the 60s, the Renaissance, you know?
L356Brian: So, yeah, you're going to have that in the cycle and it's a good thing to have.
L357Host: So do you see, do you see a, and let's just call it like American understanding of AI? I see it at like less than 10%, maybe less than 5%. And I think that people like me and you are really pushing to get things easier and easier.
L358[redacted: private, lines 358 to 360]
L361Brian: Yeah. It's, we're early, man. Like, we're way early.
L362Host: But so you think what's going to happen is that we're going to push forward. The frontiers will always be out there. Like, so today, you cannot run a household off of a local GPU.
L363Host: I think that the GPUs are going to advance. I think the open weight models are going to get better and they're going to push in together.
L364Host: But if you show somebody a frontier model right now, and then I put you, Brian, on an open code with the Quinn 3.0 whatever on a 12 gigabyte GPU, it's going to be s***. You're going to be like.
L365Brian: Yeah.
L366Host: What's that?
L367Brian: I dabble in, I've got a couple local models.
L368Host: Yeah. And the reason that you're doing that, I don't want to put words in your mouth. The reason you're doing that is you want to stay, like, you want to understand it and you want to watch it.
L369Host: And you're shooting towards that frontier and you're working on memory with a little bit of an eye back to go, where is that local? How is it doing?
L370[redacted: style, lines 370 to 374]
L375Brian: I 100% agree. And I think, listen, if you took Opus 5.5 right now and could put it in a home, I could do everything I ever wanted to do.
L376Host: That's right.
L377Brian: For the most part. I mean, I couldn't think of a daily household things to do.
L378Host: That's right.
L379Brian: Couldn't be accomplished with that kind of model.
L380[redacted: style]
L381Brian: Yeah. Yeah.
L382Host: And I mean locally. No subscription.
L383Brian: Yeah. I agree because I think it's such an interesting time, right? Like, right this very week.
L384Host: Right. Today.
L385Brian: Right, right this very moment.
L386Host: Yes.
L387Brian: You know, the models are solving math problems. They're working with drugs and new compounds. I think we're that close, man, to, like, when we, there's a lot, to me, there's a line.
L388Brian: Like, there's an edge, right? There's an edge where you get to that edge.
L389Brian: It's that roller coaster thing, right? I think we're just hanging.
L390Host: Yeah. Gravity's pulling.
L391Brian: Because we had this discussion in our Omarchy builders last night a little bit about local and, you know, versus frontier now. And then the CVEs that's been going on this week.
L392Brian: It just, it feels like we're, you know, like the, like if this next model release, it's going to be interesting. Like, I'm anxious for it.
L393Host: It's not, like, what the f*** is it going to do more?
L394Brian: Right. Right.
L395Host: But if you think about it, the skill sets are not going to get better. What's going to happen is the longer range correlation of events will get better.
L396Host: And the correlation of household events and the things that you do are going to join information together that you would never have thought about.
L397Brian: Continuous learning, like that, that's probably the next, I would think the, would be a really big jump, you know?
L398[redacted: private, lines 398 to 403]
L404Brian: It's going to be good for the consumer, right?
L405Host: Yeah. But even to the point where it's going to manage calendars. And what I mean by that is, like, if I had a meeting with you that you canceled because you had a meeting with somebody else, I had a meeting with somebody else, all that's going to daisy chain and downflow to keep everybody whole.
L406Host: Does that make sense?
L407Brian: Yeah. You know, of course, I think we all as builders think about this, too, but the privacy thing is a big concern. Like, it's a huge concern.
L408Brian: And that's why your point, local is going to be, like, local will win. If somebody can put that right package together, that absolutely would win because you don't want all of these companies knowing everything you do.
L409Brian: And that's, you're not going to get around it with what you just said. You know, you're all going to have your own APIs.
L410Brian: You're going to buy groceries. They're already doing it now, but you're going to be shopping.
L411Brian: And more and more businesses are going to serve those APIs to your models, right? That's a big, that's a big gap right now that I think a lot of businesses are just totally missing.
L412Host: Yeah.
L413Brian: Why not build a back end for these agents right now? Why not be that, you know, that first one? I think you could start capturing and be on the forefront of that.
L414Host: Yeah. I also think that when you put your own API, like the Brian Jones API out there with your MCP that represents you in public and schedules meetings, I think stuff will come up in the marketplace and your MCP will go, "Hey, Brian, you've never done this before in work, but you got the ability. Do you want me to bid on something that you've never done before?"
L415Host: And you're going to go, "Let's go for it." And then that'll become a new revenue stream.
L416Brian: Yeah. Yeah. It's interesting, man.
L417Brian: It's interesting to, like currency even, you know, like how, I just think it's going to change. I think the whole, I think we're going through some rough times for a while.
L418Brian: That's what I see. I think it's going to be rough for a little while.
L419[redacted: private, lines 419 to 434]
L435Brian: Well, it's just, that's what I'll say, man. You can never learn this all. There's just, there's, because you can do so much, there's so much to learn, right?
L436Brian: And, but I mean, I'd, listen, it's, it's been a grind. Like I said, I, you know, I work 10 hours a day, you know, and then I come home and, like I said, watch a show or two and hang out.
L437Brian: And then I come in and work and I, I mean, I've been doing it every night. Like, there's probably a handful of nights in a year and a half or whatever it's been that I've not been in here working.
L438Brian: And now working from the phone and, you know, I can do it during the day, you know?
L439Host: Right. And when you take a break, you feel like you're being lazy and you're falling behind.
L440Brian: Yeah. So I do, I do understand that. I do understand, especially if I was sitting here all day long.
L441Brian: Yeah. It would, you got to get outside, man.
L442Brian: That's why, like I said, I love the mountains, right? I love the mountains and you get outside and we took a drive this past weekend on the parkway and it was awesome, you know?
L443Host: Man, I watch these YouTube videos of these guys going up into Alaska and building a cabin. And I would love to have this set up in a log cabin with Starlink with Amazon access.
L444Brian: I agree.
L445Host: Just be done with it.
L446Brian: I agree. I agree. That's kind of the thing, you know?
L447Brian: You want to, I want a foot in both worlds, right?
L448Host: Yeah. Yeah.
L449Brian: I want nature. I want nature and then I want tech.
L450Host: Yeah.
L451Brian: So.
L452Host: Well, man, I appreciate you spending so much time with me.
L453Brian: Thanks for having me.
L454Host: Yeah, man. I'll shoot you over the link. Is there anything that you said that you don't want to be published?
L455Brian: I can't think of anything.
L456Host: All right. Are you okay with.
L457Brian: I'll put it on the book.
L458Host: Are you okay with having, so my AI will pull out about six shots, like short videos of your best topics? Is that okay?
L459Brian: Yep. Yep.
L460Host: Sweet. I will send you a public link that nobody has. You can review it.
L461Host: It's open to the world, but nobody can find it because it'll be just a random link.
L462Brian: Right. Right.
L463Host: And I'll let you look at it and then we'll put it out and we'll promote the s*** out of it. And I can't believe you're RidgetopAi. I just can't.
L464Brian: Yeah.
L465[redacted: private]
L466Brian: Yeah, for sure.
L467Host: I'd appreciate talking to you every couple of weeks or more.
L468[redacted: style]
L469Brian: You know what I'm saying? That's why I love that group.
L470Brian: It's a pretty good group in there. So thank you, [host].
L471Host: Take it easy, brother.
L472Brian: Take it easy, brother. All right.
L473[redacted: private]
L474Brian: You too.