*Posted on: April 23, 2026* *Tags: blog* ### The Death of the Craftsman As I look around the world today, the evidence of craftsmanship is plainly apparent to me. Visiting cities with rich history and architecture, walking through antique stores, or touring houses built pre-1930 all give me undeniable proof that there are people who care deeply about the things that they build. Or at least there used to be. I am finally at an age where purchasing a house seems to be a reasonable idea, and touring houses has been pushing this thought further and further to the forefront of my mind. When I look at these old homes, some of which have stood for over a hundred years, they have an undeniable amount of character. The builders who constructed them, the architects who designed them, and the people who purchased them, all seemed to have a core set of values around what a home was meant to be, and how it was meant to be built. When I look at these homes, they seem to scream at me what those values are. Quality. Elegance. Durability. You can feel the care that the craftsmen put into every inch of the house, from the intricate designs on the handrails to the minute details on the molding. This is a level of depth and detail that is hard to fake, although the buildings of today certainly try to do so. It is for this reason that I am rather drawn towards the idea of purchasing an older home with "good bones" in comparison to buying new, where the values of the builders are equally apparent to me, but significantly more disgusting. Where old houses have character, direction, and uniqueness, new homes all feel fundamentally the same. You get the feeling that there was no true craftsmen involved in the building process. Where old homes were designed and created, new homes are simply specced, and manufactured. Production as close resembling a factory line as possible. No thinking outside the box, no elaborate details designed specifically for the people who would be living there. There is no soul in any portion of the process. Taste has been traded for convenience. It seems nobody is concerned about a house that lasts long enough to one day house the family of their grandchildren, and are only concerned with having the "most" that they can afford currently, with no regard for the future. This shortsightedness, this utter lack of care for the future, has permeated throughout the production of just about everything I can think of. Ask any mechanic and they will tell you cars today are not built to last as long as older ones, yet are filled to the brim with so called "luxury" that you can't look anywhere in the car without seeing a fake carbon fiber panel or a digital screen. Clothes are cheap by design, so people care less about their presentation, and have no issue with a ever rotating closet of disposable apparel. How often around you do the people you know "upgrade" to the newest phone? This observation is nothing new. I'm sure your grandparents had often complained that "things just aren't like they used to be" with a similar level of distaste. This concept, however has started to permeate into a field previously untouched, and one I care deeply for, which has me wishing to explore this cycle in more depth. ### Personal Impact With the cultural explosion of the LLM, many have proclaimed that "software engineering is dead". While I think, looking at the merits of LLM's for what they are, that is a deluded statement, looking back through history, I can only imagine the craftsmen of the past felt the same way when the new methods of developing their product came about. After all, why would people want something that isn't built to last? Why would people choose the inferior product? Even if it is cheaper in the short term, knowing it would be more expensive in the long term? Despite these valid questions, that is exactly what happened. Once again examining the case study of the death of artisan houses, there were a variety of contributing factors that went into cultivating the current meta of building practices. Whether that be economic pressure, personal preference, or some culmination of other factors, it is undeniable that the result is that people chose the cheaper option. The more people who choose the cheaper option, the more builders lean towards providing only that. After all, why would they build something better, that takes more time, energy, and money, if they can make the same money from something worse? While I think I am unqualified to pinpoint what started this death spiral, the fact that it has happened in undeniable. LLM's have provided that same trade off discussion to be applicable to the programming space, much to the chagrin of people like me. Diving into the internals of how these LLM's work illuminate the fact that much of what makes programming, programming (such as the wrangling of complex ideas and problems into an elegant and extensible solution) are area's that the clankers simply fall short on. They can give you things that maybe look and seem like the correct solution, but are at the very best a prediction of what the solution *might* look like. For any programmer worth their salt, this is not good enough. You need to deeply understand the problem space, the technology you are using, the trade offs between solutions, and just exactly what the acceptable trade offs are for your given solution. Even with the longest context lengths in the world, the LLM's will still struggle with these types of problems. Throughout the process of writing this paper, the full version of [this](https://aphyr.com/data/posts/411/the-future-of-everything-is-lies.pdf) paper was released, and I resonate with the overall message. I can not really think of a time in human history that concerns about the impact on society in the future has ever deterred us from pursuing new technology. For the most part, this has been to the benefit of all, however its not hard to think of some exceptions to this rule, take your pick of weapons systems and there is probably an argument that the world might be a better place if they didn't exist (although I concede that there are benefits to them existing as well). My real fear is that we have no mechanism to hit the brakes on this train, and in our day to day lives more and more people will not only start to use these "tools", but will start to rely on them. We are already starting to become aware of the impacts of these tools on our learning ability.[^1] In terms of software, it means that we will have more buggy, less performant, and worse software across the board, which is astounding considering how bad the quality of most software was pre-LLMs. This is in and of itself a bleak future for me, with my investment into crafting quality software, but this could also cause real world impact to people in ways I don't think they expect. Oh, your flight was overbooked because the AI was optimizing for profit with no regard to average cancellations? Don't worry, just talk to our support team (another clanker) in order to get your issue resolved in a timely manner. Good luck catching the next plane with any degree of speed, let alone getting the compensation you deserve. Even worse [we already see lawyers using this slot machine technology in ACTUAL COURT CASES](https://www.damiencharlotin.com/hallucinations/). Ever wanted to roll the dice on if your going to jail? Don't worry, the future is already here old man. ### Moving Forward In general, I am an optimistic person, and I would like to think that we will find ways to adapt this genuinely useful (in some cases) technology to be more beneficial across the board. At the very least I hope we find better ways to mitigate the nondeterministic nature of this technology. However, I can't help but worry for the next generation. What is the world going to look like where schools do an even *worse* job at educating the youth, with no way to crackdown on usage of a tool that rots your brain? What kind of hellscape will we have to witness when LLM's start drafting legislation? These are questions that I would rather never know the answer to, but i frankly see no possible path at the current trajectory where it doesn't happen eventually, and probably sooner than we would like. At the end of it all, these are circumstances beyond my sphere of influence. I will continue to build real skills, learn deeply, and try to position myself in the ways that give me the best outcomes for myself and my family, and work hard to build real connections with people who share my values. I have no other choice than to try and craft the world I want to be in, even if it ends up rather small. [^1]: [This](https://arxiv.org/pdf/2604.04721)paper has really blackpilled me