I'm Building an Empire with AI Agents (here's the model)

You've got skill gaps. I've got skill gaps. It's time to learn how to traverse them. Oh, also I tried every single AI tool for coding and then also figured out cursor context from one of the developers at cursor and a lot more. Let's cover it. This is a daily upload and on this channel we do quick news. We do a little bit of strategy. Keep it short and snappy. So, first off, a friend of mine named Brendan, he put a three-hour video out, sorry, Brandon. Woof. He put a three-hour video out on the agent development kit, and he worked at Crew AI for a long time, and he's a developer, a great developer, and I'm excited to watch this because he knows what he's talking about. So, if you're interested in this stack, then it's definitely worth checking out. I think that this is the future. They are built to be deployed. They have Google backing them. They've got it all. So, highly recommend watching that video cuz I don't even have to watch it to recommend it cuz I just know it's going to be good. Second up is Apple is teaming up with Enthropic, which is great news because both of them have dev communities who love them, which almost doesn't feel deserved on the Apple part just because Xcode hasn't seen much. It's crazy. But if you go talk to anyone at any major company or just read stuff online, a lot of the iOS devs just love it. and it is native and they did just see the 30% tax get deleted for the US group anyways. So, seems like a great time to be an iOS dev. Frankly, got me excited because they're going to be building stuff for Xcode, which is great news. Up next, we had a really good call yesterday on the Vibe with AI group. We do one a week and it was awesome because we have some just 10x engineers now which is super cool. I wanted to just talk about this because I have three videos that are going up on my main channel that I think are worth watching depending on your cohort. But basically I'm building out a product called Echo and it's written in Python and I just I'm not great at Python but I need to be great. So I just talk about how to identify the skill gaps and then the habits around getting through those. And then I also talk about in another video just my journey going through testing every tool for that workshop. I always say that you should not tool hop. But the gist of it is that this is it. Use one or two tools. Get really good at prompting prompt chain it yourself versus not. and that every tool pretty much maps to one of these steps in the software development life cycle and that depending on your use case and where you're at, you should lean towards the top, maybe not them. Then I go through a cursor context article that's written by the guys over at Cursor. I find it's a little bit thin honestly, so I add to that in that video and I think it's good. And then I have some questions that came in. So one of the questions was, "Hey, look, you're part of this company that did well and you're doing startup stuff. Like why would you make a community? Why are you on YouTube?" It's part of a strategy that I'm running. So there's a bunch of companies that I studied and I wanted to address this. So one of a couple of them is just Blackstone and Accenture. I read the book that's from the it's a biography by the founder of Blackstone and they crushed they're currently crushing Accentra does this as well where they have software division they have training division they have services they have all these things but it's called a multi-legged stool and in my case I've wanted to do this for a long time but as a one-man band which I want to keep it you couldn't really do it but I really think that's what gets me excited about the work that I'm doing is it's not really one man band if you have a hundred agents that are doing stuff for you. So, I'm just learning how to build them and learning how to do agentic workflows such that they're good enough that they can be turned into agents because I could technically you can go and make agents right now, but they're not going to be good. They're not going to be successful. And so, that's a big reason why do multiple things. And then I also selfishly I live in a tech desert. So I want to surround myself with like-minded individuals because you are what you think. And in my case, AI first SAS is one leg. AI first marketing is another leg. And then the paid community is about the community, the events, the people around you. Eventually having different offers within that where the first one would be the online aspect of it, which is low ticket. It's what we're doing now. It's $50 a month. It's going to be 100 soon, but it includes a bunch of content of things that I've made. and then you get updated stuff on school and then you get access to a discord. So that's the online element of it. Then the second part of it would be in the future some sort of marketplace where I can go to companies and figure out how to get deals on the different tools and softwares that we use because it basically is a benefit for them. And then the members pay for their membership by getting that. That'd be the marketplace. And then the third aspect of it would be deepening the relationships with people in that network by having inerson events. So that's down the line, but that's what I think makes the most sense. And instead of maximizing profit because this is low ticket, I'm more interested in like the long-term benefits of this where network matters a lot, relationships matter a lot, then you're nurturing those over years, not days. So, another example of this is Amazon at the largest scale with e-commerce, cloud computing, and logistics. And then this is the Walt Disney one, which is really cool. If you are in the community, if you've seen this before, sorry if you just your eyes got glazed over, but I wanted to explain that to other people that are on YouTube and they're asking me about it. The last thing I want to cover is leaving Nex.js. So I a little bit of a story but so my coding journey I'm 32 now crazy and right now I'm using tanstack query so tan stack router tanstack query all those things tanstack router or tan stack start and that's after doing a lot of different things and I'll explain why I like that but it's basically if I go way back in time and it's I don't know I think I was like 18 or something and I just wrote like basic HTML and CSS to figure out what that was all about and then at 23 I was like what's this JavaScript thing tried learning that failed move forward tried this same thing again at 27 right and then my stack that I chose was like somewhere in between here was basically I think right here actually was when I was like, "Oh, this next.js thing." It was like super early. There's the pages router and just built some basic stuff with it. But then it came time when they started just introducing stuff. I don't know, like every month. And I get it. It's awesome. They're trying to make things better. The draw to it is there's like a lot of content around it. And it does look really good. And I'm not This isn't like a hate video at all cuz there's plenty of those out there. But I just found it tough because there's a lot of terms that are in it that aren't transferable. So if I learn a different framework, if I just learned React, it wouldn't make sense. But if I learned like Remix or Tanstack, the patterns to me just make more sense. And there's stuff that would be more transferable because it's more based on the primitives of the web rather than little like the new I don't know PPR is a good example. There's just a lot of those and it becomes a little bit annoying when you're managing and building a lot of different projects because then if I let's say if I build something and we're at month one and I build it and then I come back month six and I try to run it again or even just update it, even if it's not going from pages to app router, then it just breaks and I just I keep running into this and it's really frustrating And I've dealt with it for long enough. Finally, I've taken the Kool-Aid on this stuff where I follow people and they talk about it. I'm like, "Oh, they're just being haters." But there's a lot of validity to it. Even if you take out like the bashing on the triangle company, it's I don't know. It's just true. And there's other stuff, too, where you just you go and you use it and you're like, "Wow." I don't know. First of all, it's quick. The DX developer experience when you're running it, it's just bananas fast. And I'm seeing people talking online about it and they're like, "Yeah, we did benchmarking and it's three times faster." And then the predictive building is a big one cuz I was building map as my SAS and I was originally building that and I didn't realize that it was going to cost me like $20,000 a month to operate it on Verscell if I were to achieve even some usage because of the things that it required. And so there's just a lot of reasons. I don't want to get into it too much. I feel like I've talked about it too much, but I really am excited about this move. And kind of like the last straw for me was just having the same thing happen where it's okay. Like literally nothing's changed. I just haven't trust this project in a little while. And then I go to work on it and it's okay. I'm sick of it. I'm not doing this anymore. And so I'm excited about that. If you guys have any comments on that, I'm actually curious what you guys use, the viewers. If you could just drop a comment below, I just want to know what you guys use. Honestly, this is something that I'm just genuinely curious about. If you liked the video, make sure you like it. That would be awesome. And you should definitely check out the community if you're interested in it. We're always up to some cool stuff. There's typically people up here in the channel hanging out. It is a Saturday though, so maybe not. We had a good workshop yesterday. It's half off right now until we launch our new platform. Again, you should go watch this video if you're interested in agents and subscribe. And if not, you guys have a good day and I'll see you in the next