161–170 of 237 posts
If humans are replaced en masse by AI agents, companies will save billions until there is nobody left who can pay for their products and/or services because we're all destitute and on the dole. I'm scheduled to retire in about 9 years. I'm not sure I'm going to make it that long.
I know things are bad, we all know things are bad. But first you’ve got to get mad. https://www.youtube.com/watch?v=ZwMVMbmQBug Here’s the thing: We’re in the complaining stage. It doesn’t hurt enough yet. When it does, we’ll stop complaining and do something. Complaining is a good first step! That’s how it starts.
Earlier quoted context omitted.
Curiously, and this is probably just me, but I find I need to work in order to enjoy (value) my leisure time. Odd, I know.
OK, then work. If I enjoyed 5:30am wake-up times, I wouldn't require others to replicate it.
https://careers.salesforce.com/en/jobs/jr307625/software-eng...
This was one of the top jobs listed. Of course I have no idea if they are hiring people or not, but it’s at least listed.
Many commenters say that AI is overblown and they're right. But the rise in labor precarity in almost all sectors is real. If the effect size isn't attributable to AI, then it must be the case that we have been in an unacknowledged recession for over a year.
In tech, change is the only constant. From embedded software in assembly to mobile/web development in javascript to jit-compiled python. This is not easy on engineers or on businesses as they have to keep reskilling at all times. The silver lining to AI is that it is poor at reskilling itself which is why the data-cutoff is circa 2024 for some models.
I think we’re seeing the early signs of that now, as everyone giddily trips over each other to rush into the “next big thing” rather than invest into the serious R&D necessary to make real gains. We’re managing modern technologies with centuries-old methods of corporate and political governance, complete with infighting and backstabbing. In a vacuum, no rational person would have ever backed the current explosion of LLMs given the more prescient problems facing the species, yet here we are squandering water and energy on prediction machines instead solar farms, battery storage, materials science, and infrastructure replacement.
I already see personal evidence in the IT side of things. For all the bickering supporters do between VMs and Containers and K8s and Public Cloud, all I see is just different ways of packaging software with differing sets of limitations or use cases. Everyone is so fiercely competing to make their chosen tech reign supreme over all that they pay no attention to us IT folks just wishing someone would make their container or Helm chart as easy to deploy and support as a modern Linux VM might be and integrating disparate tech instead of seeking to replace and dominate their competitors. There’s a very real attitude among my peers that, despite their advantages, containers are often little more than packages with extra steps and infrastructure for their existing use cases. Whether they’re right or wrong is irrelevant, because they’re making the decisions in their employer.
Change still feels fast and explosive because folks are focusing on niche research papers, extravagant moonshots, and the various implementations of things in SV circles. Outside those, however?
It’s slow. It’s stagnant. It’s boring, because nobody actually listens to what the physical majority has to say about products - only what the biggest paying customers (economic majority) want.
Earlier quoted context omitted.
> The quality is irrelevant in the overwhelming majority of cases Software quality is especially important in safety critical applications. We should not expect an LLM trained solely on formally-verified code to produce formally-verified code. I don't think that also training on specs and hateful training material will fix that. So then we're back to the original software engineering objectives of writing better SAST…
I've been to birthday parties that employed more people than "safety critical" software development. We're talking about 99.99% of the software development jobs evaporating.
Compare traditional woodworking with modern carpentry on quality, longevity, and marginal efficiency.
From "Why Don't People Use Formal Methods?" https://news.ycombinator.com/item?id=18965964 :
> Which universities teach formal methods?
> Is formal verification a required course or curriculum competency for any Computer Science or Software Engineering / Computer Engineering degree programs?
> Is there a certification for formal methods? Something like for Engineer status in other industries?
For what it’s worth Salesforce.com is hiring Software Engineers. https://careers.salesforce.com/en/jobs/jr307625/software-eng... This was one of the top jobs listed. Of course I have no idea if they are hiring people or not, but it’s at least listed.
More at my site (in profile; no sales and low stylistic ambition), if you click on "Things I want to say" (about 1/2-way down), then "On peace amid commotion" (also about 1/2-way down), then skim that page and click at least the last link. Then read the entire page and click the links that seem most interesting.). We really can be OK despite things. (Thoughtful comments appreciated with any downvotes.)
Don't worry -- soon after that, there will be high demand for human coders as companies scramble to hire to rewrite all the buggy and vuln-ridden AI-hallucinated software. We're on the verge of two revolutions in tech, not one.
I would say that the current capabilities of genAI is like a junior dev, sometimes even a mid-level. But one main difference is that a dev is slowly learning and improving and at some point will become a senior dev and also domain specialist.
If there is a codebase created by genAI, then it’s equivalent as if all devs left the company, so no one knows why some piece of code was created in a certain way, if it was part of the business logic or some implementation detail