• I don’t think any interesting endeavor (especially with any element of creativity) is that clear cut. Software in particular sits at a really interesting intersection of goal and means. Over-emphasizing one over the other introduces a significant tradeoff.


      score 1 · canonical

  • lobsters__alanmeira@crustaceanson.theshelf.topM
    link
    fedilink
    arrow-up
    1
    ·
    edit-2
    13 days ago

    My biggest problem with the current scenario is that we have so much demand for software improvement and new software. Companies just don’t want to invest on it anymore, it is a kind of paralysis where no new great app, no new amazing service is created anymore, competition doesn’t exist and big tech is making whatever they can to protect current profit growth by enshitfication.

    If we all agree enshitfication is real, why the incumbents are not losing users and money?


    score 2 · canonical

  • There’s a lot I can relate to in this post, but I think there’s a crucial point that often gets missed in analyses like this. As software developers, we analyze this situation as a case of us vs them (i.e. management), but in reality, the management also has as much to worry about in all this as we do.

    If LLMs can completely replace software developers, then many of the companies we work for are also at an existential risk. In particular, companies that provide software services and products. Perhaps, before long, a lucky vibe coder with the right setup will pull the arm and the slop-machine-3000-ultra will spit out a great competitor. Software shops are equally afraid of losing their moats, just like software developers do as individuals.

    And even if management in a particular software shop isn’t afraid of losing their moat themselves, they will have a hard time finding investment, since the investors have to worry whether they’d be burning money on something that next year’s LLMs can just be made to write from a single prompt. That single prompt can just be “make a competitor to company X”. I actually think that as of 2026, this freezing effect on investment (combined with the brutal macroeconomic environment) might be the main factor in determining the bargaining power and the everyday experience of a software developer.


    score 1 · canonical

  • Alienation. The word the author should be looking for is alienation.

    Back when the paychecks were fatter and we could choose jobs, the benefits weighed against the risk of a stress-induced heart attack (I know at least two developers who had one before their 40s). Now not anymore.

    This is an industry in which people have always done a crappy job and called it “engineering”, with a culture primed to avoid any kind of accountability and smart enough to sell it as a perk to workers for a very long time.

    I don’t want it to burn to the ground but, oh my, how good would it be if something completely different came up from the ashes.


    score 2 · canonical

  • lobsters__toastal@crustaceanson.theshelf.topM
    link
    fedilink
    arrow-up
    1
    ·
    edit-2
    12 days ago

    Welp, I bought a place to set up our teahouse—so hopefully I can work on my own exit. I took time off to reskill to do Nix & adjacent stuff right before the LLM wave since I wasn’t happy with web front-end moving to TypeScript+React almost exclusively, just to land in this state where every, including myself is sad/angry & clinging to their existing jobs since there doesn’t seem to be a lot of prospects in the future—either bossman wants you to meat proxy, or clients are choosing agent slop. I can only imagine how much worse the Indian outsourcing is doing… I can build my own projects for fun/learning/usefulness & maybe for my local (physical) community too, but I’m not convinced that it’s going to get better before it gets worse. At least I didn’t pour a lot of my free time into doing open source stuff to have the corporations suck it up, make us pay for it, then take literally no blame or responsibility for the harm they are causing /s.


    score 3 · canonical

  • I will, day by day, try to find my flow state in strategic programming rather than tactical programming. By strategic, I mean that having the right software principles leads to better ADR(domain and no-domain guidelines), and then that leads to a good spec that agents can take over for execution. and followed by good verification phases.


    score 1 · canonical

  • I’ve long considered software development to be a modern «blue collar» job, akin to a machinist rather than a management position.

    And here we are with a new wave of technology replacing our manufacturing jobs, not outsourcing them to cheaper labour markets. Are there still machinists today? Yes, but it looks very different to 100 years ago. So it goes with software.


    score 1 · canonical

  • I also see this as the result of how anti-union the industry has been. People saw the industry as a meritocracy, where the only thing standing between you and the top was your own ability. If you ever dared to bring up the U word, that just meant you were a mediocre developer trying to benefit from other people’s hard work.

    But now something is shaking up the entire industry, and the institutions that could have helped us mitigate the pain aren’t there. This is what happens when every developer is left to fend for themselves.


    score 2 · canonical

  • I am skeptical of building this entire argument on “I keep seeing videos about this”. For one, that’s how the algorithm works; if you watch “this sucks I quit” content it will find more of it for you. Secondly, people whose jobs don’t suck and who aren’t quitting aren’t making videos about that fact; and if they were, it’s well known that the algorithm prioritizes negative content, so you wouldn’t see it.


    score 2 · canonical

  • lobsters__erock@crustaceanson.theshelf.topM
    link
    fedilink
    arrow-up
    1
    ·
    edit-2
    13 days ago

    So my current employer has us leaning pretty heavily into using code agents for coding tasks. I don’t think we’ve had enough time to see the technical debt accrue from using LLMs to write code. I have personally felt the push and pull of writing huge pieces of code with an agent, only to hit limits with what it is able to accomplish without intervention and, much worse, my lost context on the technical details.

    This is the piece where enterprise is going to start snapping back. People want to talk to people that understand the technical implementations during meetings. Running to an agent to answer the question is inefficient and I’ve seen the agent just not think about the problem properly and be confidently wrong. It’s not that the agent isn’t capable of reading the code and understanding the logic but it doesn’t have the domain knowledge of the business or entire system “in their head.”

    For personal projects, I’ve decided to put LLMs back into read-mode. This isn’t me relishing life before LLMs, any “productivity” gains are lost when you can’t navigate the code to debug or understand how the system works. I don’t always want to play telephone with an agent to learn how a codebase works. Sometimes I need to read the source-of-truth. I don’t care if people can vibe through the lifetime of a project, they will know and understand less than someone that wrote it themselves.

    A huge part of my personal and professional life is to learn new things and grow as a SWE. The project itself is not the only goal. Humans value oracles and when you outsource all thinking to a code agent, you are not an oracle.

    I’m honestly not worried about coding, neither as a hobby nor a profession. I do think code agents are addicting and it is causing burnout, fatigue, and a lost sense of purpose. I say this with confidence because I feel it in me as well.


    score 1 · canonical

    • lobsters__aae@crustaceanson.theshelf.topM
      link
      fedilink
      arrow-up
      1
      ·
      12 days ago

      This has pretty much matched my outlook so far, but I can’t help but wonder if I’m deluding myself.

      I use read/plan mode a lot, and still approve all actions on projects I actually care about.


      score 1 · canonical

  • For me, none of these larger tech companies were ever nice places to work to begin with. Which is why I spent my career trying to avoid them (though I ended up at several big players temporarily due to buy outs).

    I fully believe these terrible places are worse now. But I don’t think they were ever nice places to work.


    score 1 · canonical

  • lobsters__pyj@crustaceanson.theshelf.topM
    link
    fedilink
    arrow-up
    1
    ·
    13 days ago

    I have seen too much legacy code, with weird idiosyncratic behaviors within hard-to-inspect systems, where few (if anyone) understands the system, to believe that “AI writes all the code” will replace all developers.

    The future software engineer looks more like a lawyer to me: much fewer of us, each serving multiple clients, billing insane amounts in 6 minute increments.


    score 1 · canonical

    • lobsters__emk@crustaceanson.theshelf.topM
      link
      fedilink
      arrow-up
      1
      ·
      12 days ago

      The future software engineer looks more like a lawyer to me: much fewer of us, each serving multiple clients, billing insane amounts in 6 minute increments.

      Billing $375/hour for 3 hours work is a fast track to defaulting on your mortgage. You spend too much time finding clients and billing clients. Plus it’s hard to keep a small-job pipeline full, so you spend a bunch of time sitting around with no clients. This is fine when you’re 25 and you are fine spending a lot of time hiking instead of working and have reasonable tuition loans. It’s less fun at 40 with a mortgage.

      Furthermore, at least in major US cities, typical lawyers have been worse off than typical programmers for a while now. And lawyers are increasingly terrified of AI, too.

      So don’t automatically assume that “programmers becoming more like lawyers” is some magical land of happiness, is all I’m suggesting.

      Source: Have worked as a freelance software consultant, and once did a legal industry startup.


      score 2 · canonical

    • I think there are some pressures to smooth out that outcome: lawyers are often directly competing within a limited court throughput; for developers «being better than the other one» is not always a replacement for «adapted to our schedule», at least if the price is right.

      Only a few companies keeping developers at all could be a possible outcome. Although when all the large players go VMWare-under-Broadcom, maybe this will create a pressure that a few juniors with local LLMs are still better than the software garbage or LLM prices that you get from the Singular Corp.?


      score 1 · canonical

    • To be honest, since the introduction of LLMs, every three months, software developers have had to invent new future scenarios that somehow included writing code as an economically viable activity.

      All of this is a way to cope with the change. Some software developers imagine themselves master LLM puppeteers, thinking their collection of badly written “SKILLS” will make them viable in the future. Others stick rigidly to the fantasy that LLM-written code is “bad” and will have to be replaced “like asbestos”. If anything, it’s human-written code that will be considered asbestos sooner rather than later. Then there are those who style themselves as the WIlliam Morris of the digital age, leading a sort of popular revolt against the machines.

      All of this to avoid having the fact that sooner rather than later, most of us will have to take a job at Amazon, except it won’t be in software development.

      And in a few years, most people will not understand why our job was ever considered difficult or challenging.


      score 1 · canonical

  • folx gotta stop working FOR OTHERS and start working FOR THEMSELVES

    if you work or worked in tech, you probably have some savings. get together with your fellow coders and make some cool new think that you wish you worked on for real

    if the bosses make your life hell, quit and be your own boss


    score 1 · canonical