What we do in the age of AI

Hello and happy new year, my friends! This newsletter is going to be a little different. I don't have any tweets to share right now because I am behind on literally everything. Dozens of unread DMs, 100+ emails. I'm really sorry. All because I've been spending every second working on something. No tweets, but I do have a lot of thoughts to share, because I feel like my perspective has changed and things out in the wide world have changed. This is going to be about AI, but more broadly, it's going to be about what we do in the age of AI. For a long time I have thought that AI was moderately helpful when it comes to programming, but not as much as people wanted to claim that it was. I feel like there was a lot of hype and it didn't match up with my personal experience. That changed for me a lot around just before Thanksgiving when I used Gemini to help build out the Database School website and I realized that it was in fact quite good now. It was one-shotting a lot of things that would have previously taken me many hours. It still wasn't perfect. I still had to babysit it quite a bit. But that's when the curve changed for me in terms of how useful it actually is. At that point I feel like the scales kind of fell from my eyes. So it was both a personal change and a change in objective reality. Shortly thereafter, Opus 4.5 was released, and it was even better than Gemini. I started to get a little bit worried about the future of technical education. We've talked many times about adjusting to the game on the field. Deciding if you want to win or if you want to be right, if you want to be stubborn or if you want to put food on your family. When Opus 4.5 came out, I kicked my AI learnings into overdrive. I realized what was happening and I went pretty much all in. That's why I'm behind on everything else—because every hour of the day that I'm not with my family or asleep, I'm learning how to use AI in a thoughtful and maintainable way. I'm staying up till midnight. I'm getting up at four or five. I'm building literally everything that I can think of. And I have to tell you, it's very good. Earlier this week, Laracasts laid off a lot of their educators. Tailwind laid off a lot of their engineers. I think it goes without saying that AI has changed the game and has especially changed the game in the last two months. In the last two months I have built two full desktop applications. I've built another desktop application to about 70% finished. I've abandoned about three more desktop apps. I've built four in-browser game/simulation kind of things. I've built a SaaS app. I've built a new learning platform. By myself. One person. I've maxed out my Claude Code subscription several times. I'm subscribing to Codex, to Gemini, and I've spent a ton of money on Amp. One thing that is interesting to me is how much fun I'm having. I've always enjoyed the higher-level architecture stuff more than the fiddly bits of actually editing the code, writing the tests, setting up the CI. That kind of stuff has never been super fun for me, but I love the process of engineering systems. That part has always been a blast. With Opus, you get to just engineer systems. You don't have to write all the fiddly code. I haven't opened my IDE in two months. So where does this leave me? Slightly terrified, if I'm being honest. A big part of my income, or almost all of my income, has been technical education for quite a while. I'm not giving up on Database School. I think there's a layer of education that is still necessary. I'm still actively working on new series for Database School and will continue to push on that. But watching everyone at Laracasts get laid off and observing my own behavior, I'm not sure that that is going to be true for many more years. I think in maybe three or four years, even the lower-level database education stuff is going to die. I think that is partially because the AIs will get better and partially because the market will believe that the education is unnecessary. This is another place where I could fight against the market and try to tell everyone, no, these underlying skills are still terribly important. Or I could go with the market and what it demands and tell them, hey, here's how you use this new thing in a responsible way. I don't want to be the bitter old man that's shouting at everyone that they're doing everything wrong. To the extent that I have any influence at all, I want to be someone who is, hopefully, thoughtfully and carefully teaching people how to do the new thing well. I think there's a lot of hype out there right now, and I think some of it is justified and a lot of it is not. My personal opinions on how useful AI is for coding have changed dramatically. There's also a lot of fear out there that we're all going to be left behind. Based on my own experience, I don't believe that that is true. What I do believe is true is that we're going to have to upskill a little bit. It's funny because we, just generally as software developers and an industry, have been preaching upskilling to other people for a long time. It finally came for us. And now we're decrying the fact that this thing is taking jobs. Do remember, I've been laid off before! I know how much it sucks! My only job right now is to stay alive. And to stay alive, I'm upskilling. I'm leaving some of the things I've liked behind. I'm learning new things, which I happen to also like. I think that is what most software developers are going to have to do. I feel like Twitter is kind of the bleeding edge of people who are realizing this. Even there, we're seeing a lot of holdouts. "I'm never going to use this" or talking about how it's actually not useful at all. I fear a little bit for these people, that at some point they're going to get hit pretty hard with a reality check unless they start to gradually adopt these things. I will be doing my best to carefully and thoughtfully guide people through this transition. That's why I'm building so many things. I want to know what it's actually like to build things, to actually deploy them, to actually release them to customers, and not just become a talking head, a hype person, or God forbid, a grifter. So over the next several weeks, I'll be releasing these projects that users can actually pay for. I'll be documenting the entire journey, and I will be sharing all of this on a new AI-for-engineers focused platform. It's a wild time out there. But it's also exciting. I wish a little bit for normal days, but when there is change there is opportunity. And you better believe I'll do my best to seize it. Let me know what you think, and hang in there friendos. Aaron {{ ENV.unsubscribe\_personal | default: visitor.unsubscribe\_url | hyperlink: "Click here to unsubscribe." }}