Breakeven Brothers

You're listening to the Break Even Brothers podcast, a show by two brothers shooting the breeze about business, finance, technology, and everything in between. All views and opinions expressed are solely our own and not affiliated with any third-party. I hope you enjoy the episode. If you do, leave us a review and drop a like. See you there. Awesome. All right. Well, welcome back to episode 14 of the Breakeven Brothers podcast. Brad, how's it going? Fantastic. 2025, year of routine. You're getting stuff done. And of course, the year of AI. So yeah, doing pretty well. Pretty chill weekend. Got a lot of stuff done that has been on my to-do list forever. That was nice. How about you? Nice. Yeah, it was good. Yeah. It's 2025. It's been a good year so far. 12 days into it. But yeah, I'm on like a 17-day Duolingo streak. I'm on 10 days. What are you learning? Hi, Valerian. Chinese. Yeah. Oh, cool. Cool. That's awesome. Yeah, I heard that's like one of the hardest languages to learn. So best of luck to you. But then beyond that, been making a lot of good food. I had a nice ribeye this evening, which was nice. Had a little ribeye and watched the football game. So, yeah, it's been good. Got some, you know, I reported, I think, a couple podcast episodes ago that I had planted a vegetable garden or vegetable bed and a flower bed. And I got some sprouts coming on the vegetable garden, which is nice. Because at first I hadn't seen anything for a long time. And I was getting kind of worried. I was like, how could I mess this up? You know, just water and sunlight, right? Like, how hard could it be? So starting to see some sprouts on, I think, the lettuce and the broccoli plants. So that's kind of cool. And then a couple of the flowers are doing really well. And a couple other ones are not doing so hot. And I don't really know why. But, you know, it's all part of the journey, right? Nice. You got any technology in the mix? No, nothing yet. Yeah. No, I'm going. Like, I just threw the seeds in there. Like, I read the back of the packet and I got like a rough idea of like what planting season was. So I made sure that I wasn't planting like warm weather crops in winter. But beyond that, I just kind of threw them in there. Because it said, like, you can dig out, like, six-inch holes and then cover it with three quarters of soil or whatever. I'm like, I'm not doing that. Yeah, it sounds like a lot. These things just grow in nature, right? Like, come on. So I just, I put them in there and I was like, let's just see what happens. And so we're starting to see some progress. So, yeah, it's been nice. That's cool. Good. Yeah. Yeah. It reminds me, I think there's some new technology. Like, CES was last week or it's last week plus this week. But on my YouTube feed, it's kind of been littered with random CES stuff. And I think one of the inventions that I had seen this year was a pretty cool plant sensor that could be stuck into the soil and could measure all these plant health metrics. I'm not a plant guy. So I wouldn't really know, but it sounded pretty cool where it could tell you like if it needed water, the sunlight, all that kind of stuff. So I think for me, whenever I get my future house, trying to smart home everything. And if plants become a thing, which I think is something that comes with time, having more plants around the house, it'd be cool to stick a little mini sensor in there, get them to be, you know, watered and healthy automatically just because, I don't know, I see people buy plants and they die. It's like, you know, it would be nice to just have something that tells you what to do. They're surprisingly, I have, I have some plants that are off screen here, but there's some indoor plants kind of right to the right of my desk. And I had like six at one point, like all in different pots on like a little tiered tower running. Yeah, real ones. Yeah. And like three of the six didn't make it. You know, I watered them and get them good sunlight, but then the other three are like just completely durable because like they'll be, I sometimes I forget to bother them for like a week and they're still just growing and like they're super durable. So yeah, I don't know, like, the, you know, why some survive and some don't, you know, they're all like roughly the same. It's a couple different, like, species. I don't know if that's the right word, but a couple different, like, types I have. But yeah, like a hanging vine, a snake plant, and then I don't know what the other one is. But that's cool. I don't think they really count unless they're in your background. Like, you got to have the pristine video background of gorgeous lighting, HD DSLR camera, then you got the plant in the left and you got the bookshelf on the back. So yeah, you might need to move those to the back so that the viewers can see them. I like them where they are right here because one, they get a lot of like the most sunlight from the window, but I know what you mean. I know a good background, I see one. But then the other thing that's kind of nice, too, which I think it helps them. I don't know if this is like scientific at all, but I think it helps them is my PC is like right next to them. And I think they get some like warm air from the exhaust of the PC for like running like a heavy program or like a game or something. So I think they like the warm air. I think that's what separates out the ones that died off. I don't think they like the warm air versus the ones that are still there. Yeah. So that's my theory. I don't know. Because you know, if you go over behind your PC when you're running something like really heavy, it gets warm back there. And I think the plants that are thriving, I think they like that. That's a little hack right there. Yeah, yeah. I don't know if there's any science to that, but maybe the invention from CES could tell me, you know, hey, they like this or not. There's a really cool one that I saw that kind of blew my mind because there's some stuff that's actually useful. There's some stuff that is complete bogus and just prototypy. One that I thought was really interesting was a spoon that you could put soup in it. And I think once you brought it to your tongue or to your mouth, it would activate and make that soup or whatever's in the spoon taste more salty, even though there wasn't as much salt in it. And so it had this little sensor inside the spoon and would do something to it. I don't know what it did. And there's different levels you could activate it on, like, low, medium or high to make the food or soup taste more salty. I thought, that's so genius because sometimes I have a sweet tooth craving and I reach for sugar because I just love it. But it would be nice if I could trick my brain in a way to be like, this is sweet, but it doesn't hurt. And for the saltiness, it's not as bad for you to have high salt. I mean, I think everything in moderation, but I don't know, it was cool. I was like, that's like a really unique invention that, you know, once you hear about it, like, oh, that's cool. I wonder if that would take off at restaurants or maybe not to like save money and do less salt, but just for like enhanced flavor profile. Like, you come here, you have our soup and our spoon and that spoon actually makes a difference, which I thought, that's actually genius, you know. It's kind of creepy in a way, because I feel like I mean, it's cool. The technology is cool, but, like, can it trick you into thinking that, like, it's a bowl of French onion soup? Or is it like, like, how much can it mess up with the taste? I don't know. Yeah. Yeah. Like, you could have those vision pros on and then, like, you're, like, looking, like, oh, I'm eating a piece of steak. And it's really just, like, some tofu or something, you know? Yeah. That'd be crazy. Yeah. It reminds me of one of the cooler, this was, like, this is just completely jarring my memory of it. But there was a really cool invention. I don't know where I saw it, to be honest, but maybe it was just one of those, like, daily dose of internet happy things. I don't know. But it was like an invention where for people that have problems like feeding themselves, like either like a neuromuscular, like, the spoon shaking a lot. I know where you're going with this. Yeah, with a spoon, like it counterbalances. Yeah, the shakingness of the hand. It's crazy to watch. It's insane. You see the hand shaking, but the spoon just stays level. And it's obviously super cool to be able to, you know, assist people that need that help. So I can't imagine the feeling of, like, or at least I would hope it's this way, the complete satisfaction of, like, building something like that and like helping people, you know. Yeah, that's more useful. This one maybe is on the scale of cool. Do we need it? Probably not. But maybe it opens the door to other things. So that does really cool. I need to catch up on CES stuff. Yeah, maybe someone like wants salty food, but they can't have salt for some medical reason. There you go. Yeah. It's perfect for them. Yeah. They must have enough of a reason to make it, I would hope. So hopefully that matches something. Yeah, for sure. What else you got going on? Let's see. Last week went to the Larval SF meetup. That was cool. So I think Taylor Otwell tweeted, the creator of Larval tweeted, I don't know, maybe like a month or two ago saying there was going to be a Larval SF meetup. And for all the times that I've met with Laraville community folks or at Larval events, it's mostly been like the Larrakons, which are, you know, multi-hundred people meetups where there's conference speakers, there's this, there's that, a full show. And so I think Taylor only tweeted about it. And when I saw it, since I'm up in the SFA area, I thought, okay, I'll sign up and go to it. And I think over the weeks, there was little to no communication about it outside of, like, these initial tweets. And then the day rolled around last week and I was thinking, oh, I got the official email of, like, my invite. I don't really know who's going because the Laraville community at large is not really based in the Bay Area. Like, a lot of people that I interact with online on X, they're not really in the Bay Area. It's much more distributed within the U.S. And then even outside the U.S. is a large community. And so when the date came around, I was like, should I even go? Like, are people going to be here? Is it going to be large? They're going to be small. Like, what's it going to be? So I tweeted, I was like, hey, is anybody going to this? And, like, a few people responded, at least a few of the Laravel people who I had talked to briefly. Like a lot of new Larval employees were going to be there. And to me, Larval kind of existed at this 10-person company for a long time. And then now is in this large growth phase with VC money. And so I just kind of wanted to meet these people. And it was actually hosted at their VC funding, like, that company's headquarters up in SF. So it's pretty cool. I mean, I think it was much smaller than anticipated. So, you know, I got there. I worked up in SF that day and then, like, Ubered over to the building afterwards to go to the event. And based on how much stuff I'd seen online, I thought, oh, this is going to be pretty small. I think it ended up being smaller than they had hoped for, which to me, as an attendee, that's not a bad thing. You know, I can talk to people around there without the pressure of being at a conference where you're, like, the 50th person in the line that doesn't say something unique. It was pretty cool getting time with, like, Taylor and the other Laravel employees, like lots of folks outside of engineering, I would say, like marketing, operations, you name it. They were just there to have that DC meeting and then also have the meetup. So it's pretty cool. We had, like, drinks. Then Taylor presented on Laravel Cloud, their newest product that's coming out. I think this quarter is GA. And then after that, we had, like, dinner. And probably in total, there was maybe, like, 10 attendees with maybe, like, 12 employees or 15 employees. A pretty small chin dig. And I think to them, maybe in their point of view, they're like, maybe I won't do this again. I don't know. I'm just thinking out loud. Maybe they'll think that. But to me, I was like, oh, super cool to, like, not have to fight to get face time with any individual I'd want to talk to. Like, I could make the rounds of them some. And I think that's what people were doing. But yeah, really, really great face time. Theo.g, another, like, creator in the software space. You've probably seen his avatar. He's got, like, blonde hair. He was there, so I talked to him for a bit. Yeah, it was pretty cool. I liked it. I think it was smaller than they wanted, but as an attendee, I felt like I got a good face time with people I wanted to talk to, both, like, new people that I hadn't met, and that people I'd already met before in person. So I liked it. Hopefully to have it again. I had already been on the beta for Laravel Cloud, so I didn't have a ton of new things. But Taylor was kind of talking about the future of it and what they're working on now, which is pretty cool. Yeah, that's cool. Yeah, it's nice to have like a more, like, not intimate, but, like, smaller session, you know, every new thing. Like, the big ones are nice. Like, you know, because I remember we talked about Laracon on one of the early episodes of the podcast. And it was like a big event, you know, which has its own pluses for sure. Like, you know, just really cool speakers. Lots of good networking opportunities with other attendees that are probably, like, you know, doing what you're doing a bit more. But that's cool that you got to kind of interface more with, you know, people that are, like, working on Laravel, not, like, building things with Laravel. So yeah, I was just kind of diving into their backgrounds of, like, you know, half of these people got hired within the past 12 months. Like, where were you before? Where are you interested in? What are you, you know, planning on doing the next year? So lots of different angles of attack from the Laravel company. So I think it's a big year ahead for them. Yeah, that's cool. Where was dinner at, if you don't mind us asking? At the same place. So they had, like, it was up on this rooftop. It was great until it got freaking cold up there. So, like, it started at 5:30, ended at, like, 9:00. And from 5:30 maybe to 6:30 was, like, kind of appetizers and drinks, which was great. And then 6:30 to 7:30 was kind of Taylor presenting on Laravel Cloud and answering questions. And then 7:30 to, like, 9:00 was dinner and, like, dessert. And so dinner was out there on the rooftop, which is kind of where the previous appetizers were. Pretty good food. But again, I think when I go to these events, like the food is good, but I don't want to eat in front of people. Not that I don't want people to see me eating. It's more like I'd rather talk more. And I don't want to be, like, stuffing my face while trying to listen. It just feels kind of weird. And I think the same thing goes for conferences, too. But I told my wife when I got back home that night, I was, like, yeah, I'm kind of hungry. Not because the food wasn't good, but just because I didn't want to, like, like there wasn't any lines. I don't want to go to the front, ask for more food, then come back and then eat and interrupt the conversation. Say, hey, let me go grab two more tacos, you know, in the taco bar. And so, yeah, it was good. It was food provided all there. I didn't have to leave or anything. Yeah, yeah. I know what you mean. I am usually the same way, but I think the last, like, team dinner I had with my job, um, we ate somewhere that was so good and I was just, like, I'm not talking for the next 10 minutes. Like, I'm going to eat this, like, 100%, like grade A. It was, like, a, I think I got, like, a steak and lobster thing that was, like, you know, from, like, a steakhouse, so it was, like, legit. Um, yeah, it was amazing. So I was, that sounds good. Normally, like, I'm with you. Like, I'm going to have some light, be able to talk, not just have, like, buffalo wing sauce all over my face and fingers, you know? I got to keep it clean. But this time, I was, like, just ignore me for 10 minutes. I'm going to enjoy this. Yeah. That's the, you know, I love being out in the suburbs, but I definitely, the food. You know, my wife and I talk about, like, missing the food and a bit more of, like, a, you know, a hustle and bustling area a bit. Yeah. So, yeah, I definitely missed that out here in Phoenix. It was also, yeah. It was also my first time taking public transit up to SF to work. And then, consequently, my first time taking public transit from SF down to South Bay. And so it's a little bit of a painful train ride. It's like almost 50 minutes. And then at night, like, the event ends at nine. I need to be back home. It only runs once an hour. So I think it was, like, 9:50. And so the event was done. I, like, went to the train station. I had to wait, like, 20 minutes. And then took a 50-minute ride. And so that sounds like it would have been nice to have a car to drive back right then. But in the morning, like, getting there and finding parking at the office is kind of painful and expensive. So that part's nice to skip. But when you want to get home, like, waiting for the train at night, it kind of sucks. Yeah. And it's cool, too. How was the journey? Was there any, uh, any San Francisco locals harassing you? No, no. What I did notice is that on the Cal Train, sometimes they don't check tickets. So it's kind of like a proof of purchase system. But since the Cal Train ends in SF or starts depending on when you get on, I thought, oh, let's see what they do here before they let anybody in because there are a bunch of people waiting for the 9:50. They were checking tickets where you could even enter the boarding area. And so I thought, oh, this is probably smart. I mean, it makes a ton of sense because this is, like, the easiest time to catch people. So, yeah, they check that. But yeah, no crazy people. There was a huge line. I was, like, in the middle, and I thought, oh, I'm going to have someone sitting next to me for this 50-minute ride. Oh, that's going to suck. But it turns out there's a ton of room and I didn't have anybody next to me. So I was just kind of watching YouTube for an hour. But, yeah, it was fine. Just a little slow, I think. Yeah, I don't think I've ever taken a train to, like, like for a commute, you know? Like, I've always driven and walked at one point in my career where I lived really close to an office, which is super nice. I love that. Yeah. I remember I worked somewhere that had an office in Irvine, California, which is where I was, but they had their headquarters in downtown L.A. and downtown LA office and the Irvine office were both close to, like, the train. I don't know exactly which train it was. So people would, like, take the train from, you know, Orange County into L.A. and, you know, walk to the office and then come back or whatever. It seemed really cool. But I remember people were saying that, like, I don't think, like, Southern California doesn't have, like, the public transportation infrastructure that, you know, certainly not in New York and the Bay Area with Bart and all that. But, um, so I think they're, like, the last train back from, like, L.A. to Orange County was at, like, five or, like, 5:30, which, like, it's kind of early, you know? Like, if you were, like, working, you know, like, it just, it wasn't super convenient. So people were, like, having to, like, head out of the office at, like, 4:15 or 4:30 to go catch that 5:00 train, there being, like, that's super inconvenient, otherwise, like, you know, seems like it makes a lot more sense. But, you know, it's still cool to be able to kind of, you know, work or just do other things while you're, you know, just effectively wasting time traveling, you know. I'm definitely ready for when they do the full self-driving cars and, you know, everyone can relax and, like, check out and, you know, let us just let the cars take us there. I'm ready for that. That'd be wild. Working in, like, your Tesla on autopilot, just, like, the laptop out, you know, no cares in the world. Damn. It doesn't feel like it's that far away. You know, my kids are, you know, seven and five. And, like, I tell them, I'm, like, I think in 10 years, I don't think you'll be driving. Honestly, I think it'll be, I think you'll be taken around town, you know, with the automated car system. Waymo. Yeah, I mean, they have a lot of those out here in Phoenix. I've never taken one. I want to, but I know people that take them and they say they're amazing and, like, you don't have to, you know, deal with Uber drivers. So it's usually a plus, in my opinion. But, yeah, you know how it is. That's cool. That sounds like a nice little journey. That's pretty awesome. Yeah, that's cool. I just, speaking about code and Laravel and all that fun stuff, I hadn't, like, coded anything, I think, in a while. I think the last time I was thinking about this before we were chatting. I think the last thing I was, like, trying to do was, like, the Hello Kitty adventure guide where, like, my kids could ask Alexa, like, a Hello Kitty question and Alexa could answer. So they kind of grew out of that game, so I'm glad I didn't spend any more time on it. But I realized that was the last thing that I had, like, kind of done, which, you know, it's kind of been a while. And so sometime last week, kind of just, you know, later in the evening, I was, like, I just want to make something that, like, uses the Open AI's, like, voice chat feature. Because I really enjoy that. Like, I think last time or maybe two episodes ago, I was talking about how I've been, like, talking in Spanish with it, which has been, like, a nice, kind of, like, Spanish buddy, I guess. And so I was, like, I'm enjoying this. Let me see if I can, like, get it to, like, schedule calls where it's going to call me at a certain time and, like, you know, basically be, like, an accountability partner, if you will, of, like, hey, I'm going to call you at 4 o'clock. It's a cool idea. Yeah, we're going to talk about Spanish. We're going to talk about business or something like that. So, you know, long story short, I kind of busted out cursor first. But I have one drawback about cursor, which we can talk about. But I couldn't, like, I couldn't get into it on cursor. So I just went back to VS code, which is what I'm used to. But then, you know, as our first time, like, really, I think really trying to write code, like, with full AI now. Like, before, like, with, like, the Amazon Alexa SDK and the Hello Kitty thing, I was still mostly kind of doing it myself, but then, like, just Stack Overflow or, like, Googling things. And so it was pretty cool. I mean, I still wanted to, like, make sure I understood it. So I tried to get really specific with my prompts and, like, kind of outline it. I wrote a bunch of stuff down on paper first, then I kind of asked it these questions. So what ended up kind of being the result is it made a Flask app where it has, like, user management. So people, you sign up, you have a profile, you put your phone number in. And then the Flask app works with the Twilio API where I went over. I just bought a cheap phone number and set up that phone number like in my project. So where when I, you know, in my test environment, of course, when I, I hit the button, it calls me. I pick up, you know, it's right now it's a free version. So it says, you know, this is a free Twilio thing. Press one to continue or whatever. So I hit one and I say something. And right now it'll give me a response. But then after that, it cuts out. Like, it gives me one response. So I'm still working on it. But, you know, it's pretty amazing how to get that far for me. Again, as a non-technical, I wouldn't consider myself a very technical person. So for me to get that far with something like that. I worked on it for a total of maybe four hours. That's pretty good. Like in two, two-hour, like, basically worked on it for two days, like, in two-hour little increments. And I think I'm pretty close. And I'd give it the error messages when it was bombing out. And so I'm kind of trying to work through that. But yeah, pretty amazing, you know. And there was parts where it didn't get something right and I had to kind of go back and say, hey, this isn't working, you know, and it would go back and fix it. So it's definitely not, definitely not perfect, which I don't think anyone's advertising that it is, but nonetheless still pretty cool to be able to get something up and running and just kind of mess around. Because I remember kind of prior to, like, the widespread use of AI, you know, I was writing things in Django and it was Stack Overflow was, like, the resource. And then I had, like, a couple books, you know, like actual, like, books, you know, that were on my Kindle device where I'd, like, go look up. It was, like, production Django's. I'd go look up, like, the section on forms and mix-ins, stuff like that. So it's definitely way faster, you know, doing it this way. I'm surprised you fought with cursor. Their new update was insane. I wanted to use it. Yeah, I wanted to use it for that. This is probably a me issue, but my biggest thing with it is, like, I don't feel like I have, like, the good, like, project view. My panes are really, like, really small and, like, hidden. And I don't know if that's, like, again, I tried to, like, go in and edit the settings, but I couldn't get it to, like, look, just visually, like, where it shows all the files on the side and it has, like, my terminal at the bottom and, like, has, like, all my plug-in windows, like, where I have, like, the linters and, you know, like, I'm blanking on my other extensions. Like, I guess I can pull it up, but, like, you know, like a Docker extension and stuff like that. So, like, it just doesn't feel like home for me yet. It's, like, for someone who, like, you know, I'm used to my, like, environment. I just found I couldn't, like, I couldn't, I could do stuff in it, but, like, it just didn't feel. I felt like I was still kind of learning cursor, and I didn't want to learn cursor. I wanted to just code, you know what I mean? Yeah, that's how I felt when I first started using it. So I'm a heavy user of PHP Storm doing Laravel in PHP. And it's pretty much PHP Storm or Visual Studio code, you know, the kind of leading editor. And cursor is a fork of VS code. So it's very similar. But when I was jumping between PHP Storm, my main IDE and editor, and cursor, I was fighting because I just really liked how PHP Storm did things and the value that cursor brought wasn't high enough for me to say this pain is worth it. But their latest update has really changed my mind and I've almost shifted to use cursor more because it gives way more value and the pain that I get not having the same exact feature set, shortcuts, UI, whatever from PHP Storm, it's negligible at the current state. And I think with their specific AI models getting better and their user experience getting better, like, on their own app, I don't know. Maybe I'll do a full switch to PHP Storm or from PHP Storm to Cursor because I paid for their Cursor Pro. I think it is. And it's, like, everything in the AI space is $20. Like, any service is pretty much $20. And that's another $20 and I use it all the time. And so, yeah, I would, I think I felt the same pain you had and then I had the kind of epiphany that I can do way more here, even if I kind of eat the crap that cursor sometimes gives me that I don't love. I get so much faster code output, although I have to do a lot of directing. So it really depends on your comfortability and like how much you know you need to write versus how much you don't know. Like if I go into an area where I don't know, I take cursor's code that's generated pretty easily without questioning it. For Laravel, I know what I have to do. I just don't want to write it because it's a lot of plumbing in different files. So I describe it. I hit go on cursor. It does, like, 80% of the job. But sometimes it'll miss pretty large things. And I'll say, oh, this makes no sense. Like, please adjust these three things. And then it understands it and fixes it. But yeah, I think I've come to view cursor as like a faster me in the sense that I would get to a similar-looking code over time. But cursor just does it faster than I would by hand. So that's where I've seen the value. The UI isn't perfect by any means. It's interesting because you'd think that UI would be, like, one of the easier things to make look nice. I don't know. You know, again, like just copy VS code or copy, like, you know, I'm not used to PHP Storm or whatever, but, like, Pie Charm. Like, they all kind of have a certain look. Like, just make it like that. You know, I don't know. But one thing you were saying that was kind of interesting to me, and maybe this is my ignorance to the tool, but doesn't cursor, you say something about how their model is getting better, that's just the, you just connect to, like, Claude or you just connect to, like, Open, you know, 01 mini, right? Like, it's not their own or is it their own? I think they have both now. I think they have their own, like, cursor small edits, cursor large edits, or some, like, AI model. And I honestly believe it just wraps some model under the hood, but does a little bit more with it. I'm using Claude Sonnet right now, which is pretty much top tier coder and any coding benchmark, both by vibes, which is, you know, kind of engineers using it and by benchmarks of all these, like, software engineer benchmarks, blah, blah, blah, blah, blah. I'm mostly using that. But yeah, I think they have their own models, which as far as I know could go pretty far in the AI landscape. So, yeah, it's kind of yet to be seen. But one thing I want to talk about real quick is I know how you mentioned the AI calling you, which I thought was a great idea because I think a few months ago on Twitter, I'd saw this accountability coach where you would pay them maybe $200 a month and they would like text you once a day or text you every few days, like, how's your diet going? How's your goals going? And it'd be an honest place for you to kind of describe what you're doing or like you would send daily like food pictures of what you're eating and they can kind of shame you or guilt you into saying like, oh, you like you shouldn't be eating that. That's like the accountability. But I always thought it'd be a lot of work for them to text you on an interval or for you to text them proactively. Like, here's my breakfast. I'm going to go send a photo. But I like your approach, which is, you know, every day if I expect to call at, you know, 5:30, that, like, you know, assuming I work for myself again and be like, hey, what did you do today? Like, any challenges? What would you want to get done tomorrow? And so almost like a recap of the day. And it's low effort because you'd imagine the AI could call you. It could take whatever you said and transcribe it. And then it could store it in some, like, to-do list that evolves over time. And then it's much easier to speak than it is to type. And you can have AI commands to, like, clean up what you say. If you're saying, oh, I did this. And then that, you know, like a bunch of filler words. You can just have AI say clean it up, summarize it, extract the big ideas, put, you know, modify my to-do list, do all that. And I imagine you go pretty far with that. Or you have different templated calls. Maybe one is every day at 5:30 is work. Maybe once a week it's what have you caught up with with your friends or family? Like, who have you talked to? Who do you want to talk to? And then maybe put stuff on your calendar or whatever. Almost like a, you know, you need to get stuff done, but you're lazy. And the phone call is almost like jog your memory to do things. And the output of the phone call of your voice beckons action in some large way with your calendar, email, whatever. Sounds pretty cool. Yeah. Yeah. Geez, give away the whole plan. Damn, Brad. Okay. I'm just kidding. Yeah, so like it is, so right now it's just using, this is where it went ahead. No, it's, um, it's using OpenAI and Twilio. There's something, there's something more I can upgrade to. It's like using Twilio's say, which I'm not like a pro in Twilio's like library at all. But, you know, they have a Python library that lets you connect with your, you know, credentials, and then, you know, make the calls. You can do text, obviously. But in this case, I'm using the phone call, like, set of methods or, which I think they call say. But I know that there's an option to use, and I, you know, I didn't know this, but, you know, Chat GPT told me that you can make it better with Google's, like, text to speech. It was called something specifically, but it was by Google. And I think it was just, like, one step better than, like, the Twilio, like, original version. I think you're probably mentioning, like, sometimes how pronounce this thing maybe. It was like, you can make the transcription better by using the Google thing. And it was, like, it was, like, 200 more lines of code. And I was, like, I'm not worried about that right now. But, um, so that's something I definitely want to look into. But yeah, basically the idea, and a lot of times I think it's nice for me because, you know, whenever I walk my dog or, like, as long as I just go on a walk just, you know, periodically throughout the day to kind of just get the body moving. And that's kind of a good step away from my work and just kind of, you know, recollect yourself. And a lot of times I'll, like, talk on the phone, like, during those calls. Like, I'll, you know, call our parents or, you know, I'll call a buddy. And so, um, I was, like, why don't I just, like, call, like, you know, when I was doing the Spanish thing. I was, like, let me just call or use the voice chat. And I'm, like, talking to, uh, Chat GPT in Spanish. So, you know, when I kind of had the idea, it was, like, I'm always doing this anyways. I would have, like, scheduled it to be, like, you know, specifically, like, an accountability partner for, like, a certain goal in 2025. And yeah, basically, you know, transcribes what you're writing down, saves it to a database and then they can read back, you know, the following week or the following, you know, day whenever you schedule the next one, like, oh, how did that go? It's interesting because at some point you will, this is just maybe the cynic in me, but like at some point you will maybe question the things authority. Because you know it's not a human, rather. You know it's not some, like it'd be different if you were getting, you know, if you were getting, like, coached from, like, you know, a really accomplished coder, like just say one of the Laraville founders or, you know, because you know, like, this person's been there. They've done that. Like they've, you know, blah, blah, blah. Like just as an example. And, you know, for me, you know, having a different goal, it's like, you know, at some point, it's like, well, you don't know what to do. You're just reading, like, based on what you think you're supposed to do. You know what I mean? Like, I wonder at some point would someone kind of get fed up with, like, the responses they get from the chat box. It's, like, you're not actually, you know what I mean? Like, you don't have any authority. You're just, like, a copy of what people think. You know what I mean? I don't know. That's just, like, where my mind goes. But it's going to be an interesting experiment. I definitely want to kind of finish it out and just deploy it, you know, even just on some, you know, cheap, no-name domain, domain that I just use. But yeah, it's cool. Yeah, something I'm interested about. The one thing, real quick, because I know I've done a lot of time on this, but the one thing I want to mention about, you know, using chat or using AI to code was, I noticed that Chat GPT, you know, and I have Claude. I just didn't use it. But it gave me some really obvious errors or issues that I had to go back and fix. Like there were, like, two libraries that had been since deprecated. It's, like, how did you give me that? You know, like, you had to go back and change it. And then it was, like, it had the wrong, like, allowed methods on, like, you know, put and, like, post requests. You know, it just had the wrong ones completely. And, you know, after you kind of go back to it, it goes, oh, yeah, it needs to be, you know, pass or whatever. So that was kind of underwhelming. And I was using the best, you know, I think it's 01 mini or whatever the, one of the highest advanced reasoning one is. I was using that. And for whatever it's worth, it was kind of, it got me most of the way there. But I guess our expectations now are so high that it was, like, oh. Yeah. It was interesting. Yeah. Speaking of the voice, there was this app that I saw the other day called Yogurt. It popped up on my X feed. And honestly, being in the voice domain is very interesting in lots of different ways. One of the ones that I thought of coming out of my interviews last year, specifically the software engineering interviews. There's lots of real-time conversations, you know, like we're having right now, someone could ask you a coding question, you know, loop through the numbers from one to 100. And if it's divisible by three, print out fizz. If it's divisible by five, print out buzz. If it's divisible by three and five, print out fizz buzz, which is like legitimately a question that gets asked from time to time. I've never got it, but that is a real thing. And I'd always thought to myself, if I could write a MacOS app, I could run on my Mac, listen to the conversation, audio, like to intercept the audio, understand my voice as the person who created the app or someone who is being interviewed, or then the interviewee or the person who is, like, interviewing me as, like, an unfamiliar voice, if it could extract out what they're asking me, and then automatically send it to Claude and say, you know, I'm a software engineer, I'm in a coding interview. This is a question I got asked. Write me, like, a, you know, a solution in Swift or in PHP or whatever. And then it just, like, prints out on your screen. And funnily enough, there actually is this trending tool that does this. I saw it on TikTok. It was essentially exactly what I described. So intercepts your audio, figures out who's talking. It just has this giant pop-up on your screen that shows you what you need to write for the code. And I thought, like, if I had the idea, you know, probably hundreds of other people have the idea. And I think we're now at the age of AI apps where it's easy to write a Mac OS app given cursor. Like you might not know how to write a Mac native application, but given cursor, you can get pretty damn far with it. And two, the speech to text. So if I say a sentence, it can be transcribed so fast now that it can be in real time where, like, a year ago or even, like, two years ago, if you wanted something like this to happen, it'd be hard to write and the speed wouldn't match. And so this is, like, an interview cheating tool. I wouldn't recommend it and don't endorse it because, you know, you'll struggle when you get there on the job. But the thought in my head was for people who are really struggling or if someone wanted to be a bad actor, that could be a possibility. And then so kind of tying it back to my initial one is Yogurt was a tool I had seen, which is a similar premise. So it listens to your audio on Mac, but it's mostly useful for transcribing meetings. So if you're in back-to-back meetings, say it's, like, Google Meet, Zoom, whatever, doesn't matter what your platform is. It'll record all the participants given your audio input and output and then summarize, transcribe, do all that and store it online in the cloud. And I thought, this is, like, the safe version, what should be done. And, like, this other TikTok viral kind of cheating software engineering one is kind of the bad actor side. And I definitely want to use it soon because it sounds really helpful. Just so I feel like there's so much that happens in meetings and, like, this would probably have to be adopted at a company level. I don't think I could install something like this, like, at work to do it, but, like, transcribing meetings because so much stuff happens and, like, action items, automatic tags and, like, Google Docs, calendar, Slack, whatever. Like, I don't know why we don't have it. Maybe you guys have this. We don't have it. But just seeing, like, someone put together a MacOS app that does the heavy lifting here is, like, exactly what I would want because at the end of the day, we have so much powerful tools. It probably costs five cents to transcribe, like, a 30-minute meeting. And it would probably be done in 30 seconds. And, like, I don't know, it's just such a no-brainer in my head that it's, like, I really want to tinker with more of these voice tools because voice is, like, so powerful. Yeah. Yeah, it's an interesting point. And I'll just say it as a general comment, not specific to any, you know, certain company or any, like, certain companies. But, like, you know, because I think, like, Zoom has, like, a transcription feature. Like, they have, you know, you can record and you can turn on AI features or whatever. But, like, a lot of times when you're at, like, a certain size, like, enterprise, you get into, like, annual contracts and, like, you know, negotiated contracts for, like, you're locked in with, you know, that vendor for a year. Or, like, so many people are using it that it's really hard to switch because, you know, at one point, you know, there was a software called Blue Jeans, I think is what it was called. Oh, yeah, yeah, I forgot about that one. Yeah. Yeah. And so, he's not a meta little bit. People switch to Zoom and, you know, maybe vice versa or whatever. But, like, the switching costs and, like, time and, like, you know, if you have lots of employees, it's just, like, a lot and that's where I think, um, you know, a smaller, we talked, I think, last time about the, like, one person billion dollar exit or whatever. And I think just along those lines, maybe just a similar kind of vein is, like, smaller companies can do more and be more nimble for sure. And, like, there might be, like, a unique opportunity in the next year or two, maybe even three before those enterprises catch up where, like, you can use a new cutting-edge tool and be, like, X more productive. I mean, just what you said right there of, like, you know, getting a transcribed meeting notes and, like, having automatic action items that, like, go put it on their calendar and, like, then calls them if they're not delivering and going, hey, where is this at? You know, like, you could kind of, like, go further and further into it and maybe that would work really well. You know, who knows? And that if that gets people to do 10x more, or they're able to accomplish 10x more than what takes, you know, 10 people to do that, you know, 10 people's salaries. And, you know, it's just, you can see it making a big difference for sure. Yeah, it's like the year of the AI agents. Like, I think the folks at Google AI have really been pushing, I think, 2025 is the year of agents where, you know, someone or some quote unquote one, which is an AI agent could listen to all this audio transcribe it, do all these specific actions. And for me, as a person who builds with AI, not necessarily building any models, a more application side, building these productivity pipelines and workflow sounds really cool. But it's only useful at, you know, companies of scale. For me, if I work for myself tomorrow full time again, you know, I don't have a ton of meetings. Like, I can manage my own calendar. I think it'd be nice to have an AI agent to be able to dispatch tasks to. But where I really see this value is, you know, large enterprise, 10,000 people, 20,000 people. It's summarizing all these meetings, looking at other documents internally, because no one has time to, you know, check on internal document databases and see who's already working on this effort or, or who already did it and it failed three quarters ago and no one checked. And to me, that's, like, the amount of effort that I could do to crawl through all these documents in real time with high intelligence. Like, whoever makes that software, I mean, I think it probably should be Google, given they have the whole platform of Drive Docs, you know, Sheets, calendar. Unifying all that into a very productive enterprise workplace suite where things are done automatically and I don't need to think and it's, like, smarter than me. I would pay, you know, I'm sure companies would pay massive for for that amount of, then if you could extract that and bring it to, like, the Microsoft side of, you know, I don't know, Microsoft 365, whatever their chat app is called, Teams, all that kind of stuff, like, yeah, big, big boom there. And I imagine it'll happen by the end of this year where, like, overnight we'll just become, like, twice as productive given we don't have to do half of these tasks and we're reminded and, like, reprioritizing things consistently given this massive amount of information that AI will process. So, yeah, looking forward to that. I would like to do less, but yet do more, which is the key. Yeah. Yeah, it's pretty crazy where it'll, where it will head. I think about, um, those robots from I, Robot, you know, just, I don't know, whenever I think about, like, AI, I just think about those movies, like Minority Report and I, Robot. And then I think of Goodwill Hunting, which I don't know, have you seen Goodwill Hunting? I think I've seen, like, half of it. Oh, shame on you. It's so good. Then you won't get the reference. But there's a part where, you know, the premise is, like, Matt Damon knows he's super smart, you know, he's just super intelligent, that he was, like, a janitor. But, like, he just is a savant, basically. And then the professor was Robin Williams. At one point, I think he gets, it's been a while since I've seen the movie, but at one point he gets kind of sick of his shit, like just kind of always thinking he's a know it all and, like, just can't remember what exactly they're fighting about, but the professor, you know, Robert Williams goes, like, you may know, like, when it was, when the, like, the, like, when the, like, he's, like, you'll never know, like, what the smell of it is, like, when you step into the building, you know, and it was, like, yeah, he's, like, you'll never know, like, the humidity of, like, stepping into the room after the, I'm just making it up at this point, but, like, that was, like, what he was kind of saying, like, because you haven't been there. And I was, like, I think about that with the robots. I'm, like, you might, might know who built it. You might know how to code, but, like, you'll never, never smell what we smell. Yeah. Just kind of keep your, let them know, let them know what's going on, you know. You never been in the Stone Age where our meetings weren't transcribed and recorded and auto action items. Like, that is the Stone Age now. Never even been in the Stone Age. You know, we've been in the Stone Age as humans, you know what I mean? Like, we had to come from behind and catch up, you know, like, you guys had it easy. Yeah, you got to remind them of that every now and then. I'm excited to get a personal robot, whatever that may be. And we should put that on the bingo card for personal at-home robot for 2025. Damn. You know, I might pass on that. You imagine just, like, waking up and he's just, like, standing there. Like, I don't know why I'm giving me a he. But it's just standing there. Yeah, I'll pass. Yeah. No, thanks. We'll see about that. We should have added that to the 2025 bingo card. Yeah, that's what I'm saying. Yeah. One thing that we wanted to kind of chat about quickly was, and I think it made, I think you actually added it to our list, but I had seen it too, just being in the Python world, was, I don't want to say the release, but just, like, the widespread adoption of UV instead of PIP, do you want to take the lead on what's going on with that one, or do you want me to? Yeah, I'm happy to take it. It's been on my list for a while. So the general software community is transitioning to Rust everything. And Rust is the hottest language that's very fast. So you can think of C, which is kind of the bare bones memory, tough language to work with. So you get high speed, but then Rust brings the safety and security of managing that memory in a safe, compiled way. And so maybe a little bit hard to follow, but you can think of Rust being, like, the quote unquote successor to C. And I think we talked about it a few podcasts ago, but it was, like, kind of sanctioned by the government that we have a bunch of software and, like, legacy C and it needs to be transitioning to Rust because we get similar speed, if not better in some areas. Yet, we have much, much greater safety. So less crashes for, like, critical software. You can't write, you know, flight software in C because you're going to have a disaster with some crash or then hell's going to break loose. So maybe Rust could be an option. I don't think it is, but you get the point. And so Python has Pip the package manager. Maybe I'm not sure it stands for Python something something package. Yeah, I don't know. Yeah. Something Python package manager, I'm sure. And then UV came out. I don't know when, but it's popped up on my radar. And the first time it popped up was because I follow tons of AI folks on Twitter. And one thing that popped out since these AI coding models got so good was what they call, quote unquote, one-shotting Python scripts. And so if you know how Python works. Maybe you want to go scrape a website. You need to Pip install beautiful soup. Beautiful Soup is, like, the go-to HTML parser for Python. And what UV can do, which is my first introduction into it, was these things called single file scripts. So this guy that I follow on Twitter, he said, hey, stop using Pip for Python for your AI queries. Here's a prompt that you can use, which tells Open AIS, Chat GPT or Anthropic's Claude Sonic, that says, hey, use UV and make it a single file script. And what that does is you don't need to do any Pip install. You don't need to do any Python activate virtual environment. You can do UV, which is kind of the, you know, Python 3 runner. So you say UV and then the Python script name. And I believe if UV will look at this comment that's at the top of the file, if you do a single file script, that comment will dictate the dependencies that are required. So you can just run UV, like, MyScript.py, and if AI generated it and put in the right dependencies, it can run without any setup. But before, if you ask, you know, Chat GPT or Claude to generate you a Python script, it might say install beautiful soup, install requests, install, you know, X, Y, and Z. And then you can run this. Now, with UV, on top of all the features that it provides, which is, like, safety, security speed, which is, like, a 10x improvement over Pip. So if you Pip install, now you can, like, UV install equivalent and it's much faster. But again, the main thing that drew me towards it was single file scripts where I have so many one-off things that I would love to automate like just everywhere in my life where I need software to move data around these systems, but I don't want the headache of setting up a virtual environment. So I think UV is a successor to Pip. It's faster, you know, safer, smaller. Does all these extra features. And I haven't used it yet, but I am very close to adopting it for some stuff that I'm just kind of hacking on on the side. So once I adopt it and install it and try it out, I'll report back, but from the initial findings and documentation so far, looks very promising and it has everything that you ever need. So it seems like the natural successor, but it'll take time. Python has a massive community and people are happy with how things are, but, you know, over time as more tutorials use UV, and it'll definitely be adopted. Yeah, it's interesting because I had seen it on Twitter. I think the creator, one of the main contributors, was kind of promoting it as he should, right? So I was looking at it, and it looks pretty interesting. But one of the things, and again, not from a less technical perspective, is if you look at it on GitHub, it's like 98% written in Rust and 2% basically written in Python. And, you know, Pip, on the other hand, is 99.9.9.9 percent Python. And so I guess the success in the sticking power of UV seems like it largely is, like, based on if Rust will stick around, which, like, is a relatively newer language. I mean, definitely, you know, like you said, it's definitely a hot topic. And it seems like there's definitely a need for it. You know, it's some of these languages are so old in a good way where it's, like, they're proven to stay. Like Python's here to stay. You know, JavaScript. Well, I don't know. I predicted last time that JavaScript would go away. But, like, you know, Java, you know, like there are some that have a real staying power. You know, I don't know if Rust has, like, proven that yet. But, like, what do I know? But the other thing that's interesting with UV is it kind of, I think I read in the documentation was, like, it, I think it either defaults or, like, auto-forces, like, a virtual environment. So, you know, when I set up my program, and it's interesting, when I did Chat GPT, it still did Pip in just a regular virtual environment. But with UV, I guess it forces, like, you have to have a virtual environment, which I think is a good thing. It's not, like, a bad issue or not, like, a negative thing. But just a difference between, you know, Pip. And there was still, like, a Pip, like, usage with it. Like, you, like, still some of the commands, like, in the documentation, it was, like, you know, it's, like, UV Pip sync or whatever. So I don't know how that interplays. But yeah, it seems like people really, like, there's a want for it for sure. One of the things I remember a couple of years back, I think there was another package manager, I think called Poetry. And people were really hyped that one up. And then it kind of something happened to it. I can't remember exactly what happened. But it, like, went away or stopped being supported because of, like, you know, open source issues with, like, the contributors. I can't remember exactly what it was. But that was one that I saw. I was, like, you know, a lot of the, like, tutorial books I'd mentioned earlier, like, were, like, pushing Poetry. And then I don't, I don't see that around anymore. I haven't heard of that. I think the UV really encompasses, like, multiple tools. So not only can it do, like, your general package management, but it can do other things. Like, there's Pipx, Pip tools. I'm just looking at their docs, like, virtual ENV. All that's kind of tucked away into one thing. And then it, you know, supports Mac, Linux, Windows. Yeah, faster speed, lower footprint. Like, you really can't go wrong with that. I think you can just look, like, you mentioned, 98% Rust. That's 10 times better than running Pip. And then I also Googled what Pip stands for. And it says a recursive acronym for Pip installs packages. So kind of on the PHP front where I think PHP was, like, a recursive, or, like, personal home page. But it's, like, PHP, I don't know. It's kind of this weird joke. But I think Pip will probably be on the way out. Probably in two years is my guess. It takes a while because people don't really want to change. Once, like, large tutorial makers or influencers in the community start using and adopting and pushing it, and then people get a taste of it, the migration is going to be well underway. So it'll be exciting for that. Yeah. I think it ends, like, a package manager. It's not, like, it's that core. You know what I mean? Like, it would be an easier thing to switch, you know? Yeah. It's just you're just telling the program what dependencies you need. Like, you know, I think it's not going to have a bumpy crossover when they do decide to switch it, you know. Yeah. And when you talked about when you were chatting with AI and it's bringing in deprecated, you know, libraries. One thing that I've been using in Chat GPT more has been this, I forgot what they called it, but it's, like, web search. I don't know if you've seen it in the UI, but if you have the MacOS Chat GPT native app, maybe you're on Windows. So I think there is a Windows client potentially. There's a button in the bottom when you type in your message. It kind of looks like a little globe, like an internet icon. If you select that, it'll do search with your icon. So Chat GPT and the AI models are trained up to a certain date. So they ingest all this document up until, you know, maybe six months ago. So if you ask it, like, what was the most recent holiday, like the Jimmy Carter Hallway or whatever that came out recently, it would not have any clue because it's not up to date. But the search does grounding where it kind of does almost like a Google search and we'll find relevant data and then answer your question. And I've been using that a lot more. It's been super helpful. Like any, you know, question where I can pull real data and AI can then process it, it's been really, really helpful. Then on top of that, I talked about it last podcast, but I haven't done it yet. And so I think I'm very much on the cusp is buying 01 Pro. So 01 Pro came out during their December kind of shipathon, 13 days of Christmas, what they called it. I think it was either at the start or the end, they had launched 01 Pro. And so 01 is their reasoning model. So when you send a message, it takes a lot of time to come back for the response. 01 Pro is kind of the evolution of 01, where they have 01, they have 01 Mini. And then 01 Pro is, like, the, you know, mega researcher one that we talked about in the last pod. And I had thought, do I want to buy this? It's $200. So it's 10X, the usual AI tool price. And the vibes on Twitter weren't that great, honestly. The consensus was Claude writes better code. I forget that I sent a message because it takes 10 minutes. And I think overall the Chat GPT application is much more well-designed for real time. So I send a message and get a response back very quickly. But when you use a 1-1 Pro, it might take 10 minutes. And maybe you had your phone open and then went to sleep. And I had heard some reports that if your phone went to sleep or you change tabs or but kind of something that you get bored, distracted, or whatever, can cancel that response output. And that has this really negative feedback loop. And that was the first consensus. Now I've been seeing a shift. And even Sam Altman called this out that I think people learned how to prompt 01 better. And I had saw one tweet recently talking about, I wouldn't call, you know, talking with L1 prompting. I forgot the phrase that this user had put it, but it was the perfect way to describe it as, like, I think he mentioned I'm briefing 01, not prompting it, where in my mind, prompting is, like, you know, write me a four loop using Laravel that iterates through eloquent or, like, makes a DB query. Like, briefing it is, like, you know, this is the goal. This is the context. This is, you know, the expected outcome. Here are some warnings. And, like, here's the relevant context. So I think the shift has come from really how to maximize output with this really smart model. And ever since I saw the kind of vibe shift, maybe a week or two ago, I think a lot more people have bought it. A lot more people have followed the general advice of, you know, briefing, which is a much longer version of a prompt, and have seen much better results. And due to that, I'm slightly influenced, I think, to give it a try. And, you know, if I spent $200 and it sucks and it's not better than Claude, then not the end of the world. But I've heard these success stories of quote unquote one-shotting pretty complicated logic and I'd ask folks, like, hey, what's the prompt? What are you doing? And I've heard mixed reactions of, oh, I just asked it, like, convert from, you know, like Python to Rust and it could do it all. And I was, like, damn, because I picked up working on split my expenses a little bit, you know, this past weekend. And I'm doing some complicated currency conversion. I really won't go into it because it's too boring and complicated. But I need a really detailed look at the code base. I need some decent math too with the currency conversion, just, like, logic checks, and I really don't have the time to spend, like, eight hours, like, a full day to figure it out. And at this point, I'm, like, maybe I can buy 01 Pro given the increased hype and attention on it on Twitter. Give it a go, give it a spin, you know, maybe spend, like, 10 minutes writing a brief of, like, what my feature is, what my app is, what I wanted to do, where the problems are. Fire that off and see if it can one shot me some code because, man, I have been really struggling with the currency conversion feature for almost, like, two, two months now where it's at that 80% mark and that last 20% is hell and it's a really important 20%, but it's not something I want to do. So I thought, what better than to offload work to the AI overlords is just, like, push me to the finish line. So hopefully by next pod, I'll have a result. But yeah, we're seeing a vibe shift in 01 and 01 Pro specifically and I really want to get to the bottom of it, whether it's a prompt change, they rolled out some update or, you know, it's actually not that useful. Yeah. Yeah, I'll be curious. Yeah, if you do end up doing it, you'll have to report back some results and, like, do your own little, you know, your own little testing, if you will. Yeah. Yeah. It'll be cool. Yeah. Cool. So I think we're almost at time. I had been digging through my, my bookmarks. And honestly, every time I bookmarked something on Twitter, I think, oh, this is, like, so good. I should go look at this. And then as I scroll through my bookmarks today, I thought, what the hell am my bookmarking? Because all these are junk. So not to say that one I chose today's junk, but the one I chose today was this Microsoft startup program. So it's, you know, I think a few companies have this. I think Amazon has this. I think maybe Google has this, essentially, like, a startup program that you apply to get almost, like, funding or small grants or free services. And the one I saw on Twitter was the Microsoft one. It gives you, kind of reading off their stats, $25,000 off your Stripe fees. So if you make money, you pay less to Stripe, 75% off LinkedIn Premium. So again, if you're hiring, that can be a huge value add. And the biggest is $150,000 in Azure credits. So Azure being the kind of compute, the Amazon Web Services, Google Cloud competitor. And that's to get your startup kind of locked into Microsoft. And then once you make good money, you're paying them a large check where they're handing a lot up front. So if you are an indie hacker out there, if you're looking to get some free money for Microsoft, definitely check it out. I'll link in the show notes, but I'm pretty sure it's just kind of like submit your LLC, you know, a few bits and pieces of information, and then probably might have to have something on the platform, like something deployed on Azure to get approved. But yeah, I think there's a lot of opportunities out there for people looking to pick up free stuff for small startups or, you know, new bets just so they can get you hooked on their platform. And this one sounds pretty good. I don't even know how I'd spend $150,000, so it's almost like a challenge to crack that. Specifically on Azure, like you have to spend on specifically on Azure, yeah. Yeah, that's crazy. I wouldn't have no idea. Cool. Yeah, that's interesting. I have two kind of bookmarks, I guess. The one, the first one is a tweet from Derek Thompson, who's a writer at the Atlantic. I didn't know that. I'm just reading it live. But he has a podcast and there's a podcast that I want to listen to, but in the tweet he kind of outlined the biggest, like, points of the podcast. And I guess they go over the 2025 economic forecast with someone who works at JP Morgan. And the four kind of bullet points I thought were really interesting and I wanted to kind of, you know, drill down into, to use some corporate speak, is first, big tech's dominance and profit is unlike anything in our lifetimes, which is, you know, I think it's one of those things that you kind of know, but it's just interesting here presented with and they have some graphs in the tweet which we'll all link. But then where it gets, I think, more interesting is, you know, second, their unprecedented cash flow is setting the stage for an unprecedented CAPEX build out in AI, which is, you know, kind of all things that we're talking about this episode. So I thought it was super topical. And the other two are more of, like, a comparison with, you know, the United States versus other global powers and basically the corporate outperformance of America versus Europe the last 20 years is unbelievable. And I had seen this like in some just discourse on Twitter. And it was mostly from Europeans lamenting how just the bureaucracy of, like, the EU just stifles any, like, motivation to innovate. And that's not to say that, like, would they want to be American at all, but just kind of saying that, like, the business climate in the EU is so much harder just due to all the regulation and taxes and just lack of risk are, all the USBC, everything comes out of there. Yeah, look, there's plenty of things that, like, I think Europe gets right, but, like, you know, that one, you know, seems like a lot of Europeans are saying, hey, like, this is, it's hard to build here. Like, we have to build elsewhere. And then the last one is, uh, China's economy is slowing. So, you know, I feel like I've read that a bunch of times. But yeah, just overall interesting, uh, so I want to listen to the podcast, but interesting kind of hook on the tweet, definitely caught my attention. And then they include some good graphs kind of showing the four points that we mentioned. So my first one. And then my second one, it's not more of a, it's not really a link, but it's a reading list article from Jason Stats, who I've mentioned a few times on the podcast. He's a very influential accounting person. I don't know exactly what to call him. But he listed out a bunch of books for accounting firms, accounting people. And a couple of them I have and I've started to read. So the two that I have that I've either read or started to read is Atomic Habits, which I think everyone has at least heard of that book. Then the other one is $100 Million Offers, which I think I mentioned on this podcast, not that long ago. So I'm, like, halfway through that one. But then he has much of other ones. The Accountant Marketer, I do have that one as well. I am reading that. So a bunch of good books that I want to check out. My goal is definitely read a bit more this year, specifically, like, books about, like, you know, kind of just running a business and, like, just trying to kind of understand more than just kind of, like, what I do day to day, you know, kind of, like, sales and marketing. And we both have lamented our marketing skills. So just trying to kind of get better in different areas. So I'm going to kind of take a look at that list. And one of the books actually, the Four Conversations is, I'm in, like, a book club kind of thing and that's the book that we're going to read. So I haven't read any lick of it yet. But I think that one's going to start reading that at the end of January, I think, in the book club thing. So I'm excited about that one too. It's all about sales and stuff. So, yeah, I'm excited for that. I took a look at the post that you linked. I love how he starts it off on LinkedIn. Can you read? I can. You're my favorite books. Yeah. He's really nailed his, his style for sure. He's got a lot of personality for sure. Yeah, yeah. Yeah, he's been a great resource for accounting, I think, as a whole. So yeah, he's just got good stuff, good stuff posted. Yeah, I'll link those in the show notes. And yeah, if anyone's interested, they can check those out. Cool. And then hopefully before the next pod, yeah, I can have an update on 01 Pro because I think even talking about it gets me excited to bite the bullet and try to eke out all the value I can. I think even Sam Altman says they lose money on the $200 a month, where in my head that seems kind of nuts. So we'll see if I think the value is there. Yeah, I don't know if I believe that that quote, to be honest, but, you know, whatever. Yeah. It doesn't influence me. You can still do whatever you want, you know, I guess. But yeah, yeah, I'll be curious, too. I mean, that's a, that's a big chunk of change to spend. But it is, you can just, it's month a month, right? You can just cancel. Yeah, yeah. I would do one month max for sure. Because if it was super working, it would be an easy continuation. But even then, like, I think I need it on and off. Nothing that I'm doing is really that hard. This is more just, like, a specific problem. I think it could be useful for it is a good use case for. So, yeah, I'll report back. Okay. Cool. Awesome. Well, I think that's a good wrap on everything, Brad. So yeah, we'll get everything linked and bookmarks and all that kind of stuff. But yeah, all in all, another good one. Awesome. Sounds good. We'll see you next time. Yep. See you next time. Hey, hope you liked that episode of the Break Even Brothers podcast. To find more of episodes, you can go to breakevenbrothers.com or find us on YouTube, Spotify, Apple Podcasts, or Transistor FM. To connect with Brad, you can go to BradleyBernard.com or at Bradley Bernard on X. And to connect with me, you can go to BennettBernard.com or at the real tech CPA on X. You can also find us both on LinkedIn. Thanks. See you next time.