I’m not an AI lover or hater. I find it a useful tool for some things but get really annoyed at it being crammed down our throats. AI "art" is about the dumbest thing humans have come up with, but I do find Claude helpful when it comes to helping out with coding (with the understanding that I still need to know what is really happening). The economic and environmental toll is bad but feel like both could be managed with some cooler thinking. It is absolutely overhyped, but unlike Cryptocurrency and NFTs, I think there actually are use-cases for it. I'm not writing this as a way to convince anyone of anything, nor of playing the role of "enlightened centrist", but more of a glimpse into how I'm seeing and handling this thing that is difficult to avoid.

At work we got the mandate that many are getting that we need to include AI in our workflows. It isn't a super high pressure thing like a lot of the big tech companies at the moment (wtf? tokenmaxxing?), but it is definitely being pushed. Since that is the case, I've gone all-in on building a workflow built around an agentic team for building out some of our infrastructure code and surrounding pipelines. After some time learning about how the models work and setting some good instructions/guard rails, I am happy with the results. I have based my workflows around patterns created by my team and I, so I feel confident in my understanding of the internal workings of the modules. I can still spot when Claude does something goofy, add the antipattern to my documentation as something to avoid, and be fairly confident that the issue won't come up again. It has left me with a surprisingly effective workflow that has allowed me to implement things I didn't had the time for in the past. However, it has also given me distance from the nuts and bolts of the code since I am now focused on managing a workflow instead of the code itself. My terraform-fu is most likely starting to corrode, so who knows how much longer I will have the same understanding of my patterns. I don't know that I care a whole lot. I think the reason why dovetails into another aspect of my life.

Like many of you I'm sure, I started work in technology because I love it. Also I'm sure like many of you, I got the advice to "make a career out of what you love." It turns out, "If you do what you love you will never work a day in your life" is bullshit as a worker bee. The love of technology has been squashed by decades of the grind of capitalism, long hours, and implementing things that, in the long-run I don't care about and oftentimes just get thrown away. By the time I get home/log off most nights I don't feel like dicking around with technology anymore. I still have curiosity and fun ideas running around in my head, but no energy to do it. I know this is a common sentiment among my peers. However, since I've embraced this AI workflow at work, I feel like it has freed up my brain to actually have energy for these fun technology projects that have been floating around in my brain for years but haven't had the energy for. I honestly don't know if this is a good or bad thing, maybe somewhere in between. Lately I have begun a cyberdeck project using microcontrollers like the ESP32 and RP2350 and creation of my own OS, with the goal of learning Rust. I know higher level programming languages and Python and Ruby to a reasonable degree, but have always wanted to really learn and understand a systems level language. For the first time in a long time I have the brain power to dedicate some time to it. I'm not sure how long my attention span will hold out, but I know that, for the time-being, I am enjoying taking the time to truly understand what is happening. It is more of a labor of love, so AI has a much smaller role here than at work (mostly for helping me understand concepts that are new to me). I'm sure I could build an agentic workflow to whip this thing out lickity-split, but feel like it would be cheap in all meanings of the word. Like my work code is a commodity and home project code is more about curiosity and love of computing. Oddly, this is probably a beast of it's own making, since I am feeling the push to get more out of lower-end hardware now that prices of components have skyrocketed. Now is a good time to learn to get more out of little.

I often think about the Star Trek TNG episode "When The Bough Breaks" where a people called the Aldeans had become so dependent on a computer that they stopped learning and inadvertently caused the seeds of their own potential destruction. That idea is terrifying, especially when I already see a lot of people completely dropping critical thought to depend on a computer. I think that is what is driving me in my current direction, a desire to understand how things work at a base level and to keep the idea of critical thinking alive. If we don't collectively keep these ideas going, we too may find ourselves in a situation similar to the Aldeans.