3 ms·

Tbh, while feeling sorry for the OP, I never felt sorry or compassionate to such jobs. They feel like a page of code, but it’s the same page of code every day.

by wruza 2y ago

Tbh, while feeling sorry for the OP, I never felt sorry or compassionate to such jobs. They feel like a page of code, but it’s the same page of code every day. You learn it and then just do, over and over again. Expecting a career and retirement out of this is akin to expecting it out of uber driving or food delivery. Try getting a job that isn’t a set of reflexes. The comments are right, the fact that it wasn’t automated away before AI is just an economical accident.

znpy 2y ago
The problem is that most knowledge work is “a page of code” and largely ai-automatable. We like to think it isn’t, but it is. If not exactly today, it’s going to be in 3-5 years. Will you hold the same opinion when you’ll get sacked and replaced by some kind of LLM?
- wruza 2y ago
  The problem is that most knowledge work is “a page of code” and largely ai-automatable. It is 90% automatable before AI and is not a problem, it’s progress. Will you hold the same opinion when you’ll get sacked and replaced by some kind of LLM? I’m not in the job market, but I look forward to be able to offload 90% of what I do to an AI assistant that doesn’t suck or to be replaced completely cause it will mean such transformation of societies and the world that disrupts everything I hate in it fundamentally. I’m business logic oriented type, for the most part. I only care about software because it’s hard to make. AI takes some technical parts? Good (we are sort of here if you squint). AI takes system design, visual automation, accounting, communication, integration, data analysis? Good, I can just translate client/partner ideas into it and get the results. AI started to understand my clients? Good, it means my own ideas became testable without proper formulation. AI replaced my clients and me by inventing ideas itself? The latter is a complex sci-fi question. Let’s make AI that doesn’t shit itself at the first step off the beaten html boilerplate path then we’ll talk about the prospects. Anyway, I think that these “I’m just a gear in a gearbox that has zero value outside of it” jobs were doomed from the beginning if not for AI but for purely socioeconomical reasons and aren’t worth a human attention. Backlog programmers included.

by wruza 2y ago

znpy 2y ago

The problem is that most knowledge work is “a page of code” and largely ai-automatable. We like to think it isn’t, but it is. If not exactly today, it’s going to be in 3-5 years. Will you hold the same opinion when you’ll get sacked and replaced by some kind of LLM?

wruza 2y ago
The problem is that most knowledge work is “a page of code” and largely ai-automatable. It is 90% automatable before AI and is not a problem, it’s progress. Will you hold the same opinion when you’ll get sacked and replaced by some kind of LLM? I’m not in the job market, but I look forward to be able to offload 90% of what I do to an AI assistant that doesn’t suck or to be replaced completely cause it will mean such transformation of societies and the world that disrupts everything I hate in it fundamentally. I’m business logic oriented type, for the most part. I only care about software because it’s hard to make. AI takes some technical parts? Good (we are sort of here if you squint). AI takes system design, visual automation, accounting, communication, integration, data analysis? Good, I can just translate client/partner ideas into it and get the results. AI started to understand my clients? Good, it means my own ideas became testable without proper formulation. AI replaced my clients and me by inventing ideas itself? The latter is a complex sci-fi question. Let’s make AI that doesn’t shit itself at the first step off the beaten html boilerplate path then we’ll talk about the prospects. Anyway, I think that these “I’m just a gear in a gearbox that has zero value outside of it” jobs were doomed from the beginning if not for AI but for purely socioeconomical reasons and aren’t worth a human attention. Backlog programmers included.