@aubreypwd

I am feeling truly grateful today—I do not take for granted that my ability to take a couple of months off and just quit my job is solely possible because my family lives a debt-free life. In 2016, our family paid off all our debt and we’ve lived debt-free since. Other than rent and utilities, literally everything else can be put on hold—I get to be in control. This freedom has also allowed me to save up a significant amount of money to lean on and avoid the stress of having to meet payments to anyone.

It’s all because of Dave Ramsey—who may not be everyone’s cup of tea (sometimes he isn’t mine)—but when it comes to money and hope, I cannot recommend his principles enough.

I also need to give a little credit to myself for being such a cheap tight-wad this year.

Scope creep. Sudden project changes. Indecision. Endless “iteration.” Half-baked plans. These aren’t just annoying—they’re symptoms of a deeper problem: not doing the hard work upfront—figuring out what you want!

This is why Agile has turned into the chaotic mess it is today. People show up with unclear goals, vague specs, and lean it all on their developers—then call it “agile.” But that’s not agile. That’s a project meltdown waiting to happen, and developers are the ones who absorb it as frustration and the burnout.

Lately, I’ve been reflecting on what’s been eroding the joy from my work—one thing keeps showing up: the annoyance of things constantly changing—not because they need to—but because people can’t make up their minds! A lack of thoughtfulness has created this bloated, ever-shifting “iteration” culture we’ve all somehow come to accept. But constant indecision and poor planning isn’t a sign of agility—it’s just a frustrating distraction!

This might not sound flattering on a resume, but the reality is real: the noise, the pivots, the wasted work—it all just sucks for developers!

This is where I still have hope for AI: It should make it easier for leadership roles to visualize, prototype, and clarify their vision—before pulling developers. This way developers can enjoy their work, build exactly what you want, and do deep work! Figuring out what you really want is amazing, and in my reflecting I have found well-planned out projects the most rewarding.

Small tweaks are fine—don’t get me wrong—but rethinking everything constantly?—or, gutting huge chunks of the project at the end? That’s not agile—it’s frustrating and demoralizing!

A few nights ago, I went on a long drive with nothing but my thoughts. I just wanted to ask myself a question: What would I do if I didn’t have to work? If I had a million dollars or something, and didn’t need to earn another cent—no pressure, no career ladders—what would I choose to do with my time?

Read the precursor to this first for more context: It’s time I spoke up about AI  (and return back to this when you’re done)

What came out of that surprised me—it wasn’t an answer to the question—but a realization sparked by it:

I love putting things together, and I love figuring it out.

I started looking back at the things I’ve loved since I was a kid:

• Computers
• Programming
• The Web

What I loved most wasn’t just the things themselves—it was the curiosity they sparked in me. Sure, I could figure out how to put a car together, but you have to buy all the parts. With a computer, everything felt like it was already there, just waiting to be figured out and solved. The Apple IIe my mom brought home when I was 12 was, in so many ways, a box full of virtually infinite pieces to put together and problems to solve. From the internal hardware I pulled apart and put back together, to what showed up on the screen—I found something that satisfied my burning desire to take complex things, figure them out, and put them together. All in one little beige box.

It’s 2025, and I’m not tearing machines apart much anymore—unless you count the IIe I brought home—once again— a couple years ago.  The Web back in the early 2000s gave me an infinite landscape to explore, learn, and tinker—which led to my love of coding. When I found out later in college that I could get paid to do this—shit!—I found my place.  That joy of wrestling with complexity, figuring things out, and hitting the refresh button over and over until it worked—that’s at the core of what I still love to do. That’s what still drives me. What I realized on my drive was  I had lost what I was really getting out of what I did five days a week, eight hour a day.

And that brings me back to AI again. Like I said before, I’m not against LLMs. I use them all the time—to help me figure things out so I can put the pieces together. That’s what’s amazing about them! They save me from the pain of combing through Google, forums, blogs, etc. When my mom brought home that Apple IIe, it came with a manual, something I could read and follow. I don’t think I would have done anything with that beige box without it. That’s how I see LLMs today: an incredible manual—one that sometimes hallucinates, yes—but helps me learn in the same way I used to when reading those old manuals. Speed and productivity is not what I love about LLM’s—it’s about that curiosity. And sure, it can spit out code—but that just helps me understand something better. All those years ago, I learned by reading the examples in manuals, forums, blog posts, etc. Today I use LLMs to learn more about what I’m doing better and, sure—faster. LLMs really fit in with me in the figuring it out part, what it doesn’t fit with me in—is the fun behind putting it all together.

Tools like Cursor are different. Yes, you can still learn with it—I hit “Ask” constantly when I’m digging into a complicated codebase. But Agent mode isn’t built to help you learn; it’s becoming more and more about doing it for you (AI-first).

And again, I’m not okay with that.

It doesn’t just suggest or give examples (—it can), it writes. And companies are pushing more and more for you to let it put the things together for you—that might be great for “builders,” but it’s not okay for learners.  And that isn’t a statement about how good or bad LLMs are at coding—I don’t give a shit.

Maybe I didn’t answer the question I asked myself on that drive—but it helped me remember something going on deeper inside of me.

Figuring things out and putting things together is what I do.

— 

(I’m was literally talking to my youngest daughter as I wrote this—she’s having a hard time figuring out how to quit a game—I had to remind her that she can figure it out and that figuring it out is fun!—And yes, she figured it out.)

Last night I spent some time—now that I have it—playing around with the idea of switching back to Jekyll or Hugo for my site. Well, after tinkering with both—and having issues exporting my WP content to .md—as you can see I’ve just decided to keep the current site as it is. So what’s this site running?

WordPress Classic

Yes, I just wanted to try it out. I just want to blog, I don’t need a block-editor. Just the classic editor and the most basic function: posts & pages. So far WP Classic has worked out. It’s bare-bones and I’ve modified it to be even more so. Categories and tags are disabled. It runs an SQLite database integration for easy backups, and that’s it! Comments are even disabled! It’s currently running a theme called Boxstyle with some CSS customizations. It works, but I’ve seen the code—it’s a mess. Maybe one day I will fork it and clean it up.

And yes, I don’t use pretty permalinks—not a fan of having to mess with those when you move to another platform.

Hugo is so much better than Jekyll by the way. But I already moved my content from various sources to WP and the idea of re-formatting it again for either of these platforms as markdown just didn’t appeal to me. I was going to use this theme for Jekyll which I really liked, but maybe I like this one better.

After over three years at AwesomeMotive, I’ve made the difficult decision to step away. This wasn’t a decision I made lightly. My time at AM has been full of amazing opportunities, growth, and some truly remarkable people I call friends. From taking me around the globe to the amazing autonomy I was given, I’ve been incredibly fortunate to work there. I leave with nothing but gratitude and will always recommend AM as a great place to work.

But, recently I’ve realized that the way I find meaning in my work has started to diverge from the direction that much of the software industry is going. The rise of “AI-First” has been fast and invasive, and I want to be clear: I’m not against AI, but I’m not okay anything removing joy and craft in my work.

I think DHH sums up how I feel about AI pretty clearly:

I’m using LLMs all day long, but I’m not letting it write my code… I want to reserve the fun part of programming for myself.

To me, coding isn’t just a means to an end. It’s a creative act—an art form. I care about how the code is written, how it’s architected, and the learning that goes into mastering a craft. Which I think leads to something deeper: felt growth, deep understanding, and the kind of mastery you only get by doing the hard stuff yourself. It’s not just about finishing a task—it’s about how you got there. It’s something to be proud of—and knowing when another developer sees your code, they might say to themselves, “Nice!” The finished product? That’s the result of someone doing this kind work, something built with a genuine love for what they do, not a prediction.

Code is Poetry.

This phrase has always meant something special to me. It’s at the heart of how I’ve always felt about WordPress. WordPress isn’t just a platform I work with, coding within WordPress has brought me years of joy. It’s where I discovered how much I actually loved coding, how much I cared about the beauty of it, learned more about it, and how good it feels to ship something that has my code in it. That spirit was once deeply woven into what made WordPress special. I worry that it’s slowly losing that spirit out of fear of being left behind—as large corporations gain too much influence in the open source ecosystem, shifting focus towards product instead of process. It’s sad to see, and I hope WordPress continues to somehow hold on to its original spirit…with AI or not.

The pressure to use AI for everything is clearly growing. In many places, it’s being measured and increasingly shaped by performance metrics over quality, and now the direction feels like it’s being forced into something mechanical—away from creativity—and towards raw productivity. Human coding is starting to be seen as a performance issue rather than a skill.

And I am not okay with that.

I can’t speak for everyone. Some people are fine with this shift toward speed, performance, and output. But I believe this mentality is sidelining something really important: craftsmanship. It’s not something AI offers through pattern recognition and predicting what tailwind classes to use, but real craftsmanship that comes from genuine human curiosity about what they are trying to accomplish. I hope in the future we’ll see a renewed appreciation for that kind of work. AI-generated code will feel like junk, and when that happens I hope it will be the people who never let go of craftsmanship that we’ll look to again.

Read the continuation of this: What I found in that beige box

With all that said, I’ve chosen to step away for a bit. I’m going to take a few months to think, reflect, and explore where I choose to go next in all this. Maybe there’s a place where I can keep writing code with LLMs, and by hand. Or maybe it’s time for something entirely different.

—I don’t know.

But I do know this: I can’t just give in. I want to save the fun parts—all for myself.

So TwentyNineteen just wasn’t cutting it. Just wasn’t loving it… So decided to give this one a try, again with some customizations. Hopefully it sticks around. I really would love to build my own theme, but I just simply don’t have the time (I’ve tried before). 

Finding a WordPress theme is hard (no, don’t care to build or maintain one). But I’ve settled on TwentyNineteen with some dark mode enhancements. Wanted something simple. It’s not perfect, but I will continue tweaking it to my liking.

As someone who uses Apple Notes and Reminders daily, I have to admit Apple really makes crappy software.

As a senior developer, I’ve found AI incredibly useful for boosting productivity and handling a lot of tasks. It’s like having the Internet on my computer and it can speak to me, it’s really awesome. But, it can be infuriating when it doesn’t quite get what you need, especially with more complex stuff. It feels like working with an amateur developer who keeps missing the mark over and over, needing constant guidance and re-instruction.

For me, this has led to something I’d call “AI burnout.” The frustration builds to the point where I just want to give up trying to use it. It’s annoying, exhausting, and frankly, not worth the energy. I begin to distrust it and lose confidence in its ability to help, to the point where I’d rather push it aside and I often start to continue working on my own.

I wonder if other developers feel the same way—when does the help stop feeling helpful?

Yes, when you hover over the very edge of the Safari (any in many other apps) window, the sidebar will trigger for you!

Now for many years I have had this issue. It became more prominent in my ChatGPT.app where the sidebar would somehow get triggered automatically. I thought it was a bug, but it’s actually a feature. Usually I slide my cursor to the edge of the screen out of the way and apparently I was hovering over this area since I maximize most of my apps.

Not only that you can even drag the sidebar into not existing anymore by simply dragging it towards the edge of the window and it will go away! 

Go ahead, hide a sidebar in an app (try Finder) and hover near the edge of the window. 

It’s a feature!