I am super curious how you went about the port using LLMs. At $WORK we are looking to port code, preferably with LLMs, and it seems daunting, even with a test suite. Do you have an approach that works well for you?
The question I'm left with: in the past, the uproar over these types of changes seemed to make companies change their mind when considering very anti-consumer decisions. Now, they just go ahead anyway.
What's different? How do we get back to how it was before? I know the current political climate is one that enables this sort of thing. There are parallels with the current movement also WRT to the employer/employee relationship.
Beyond that, there's still more at play. In tech, and specifically on this site, I see a lot more complicity and fatigue when discussing these issues. I can't help but think that also contributes. I'm not saying everyone should always be mad at everything. But it does seem like there's a generational component to this where we haven't passed down an essential feature of a hacker, namely the anti-establishment bent.
I suppose that's collateral damage of a culture tolerating lots of people rushing in to grab their bag of cash and then get out.
You're not. This site was also bullish on using LLMs as therapists, which defeats the very point of them, and reflects a lack of knowledge on what exactly therapists do for people.
More on topic: if the article's author arrived at a definitively negative result would this have shown up on HN?
Yep. There's also been a relentless push for the past two decades or so to standardize many aspects of software development:
* programming languages (JS)
* frameworks (React)
* open source libraries
* platforms (Web, mostly)
* design systems (shadcn for newer apps)
Guess what makes it easier for automation to come in?
Our need for it to be easy and standard contributed to the success of LLM use in software engineering. I suspect it would have done well without some of those factors, but it may have taken longer.
How I feel about programming is often a reflection of how I feel about my own internal world because it continues to be a part of my identity. Once I found a job that valued my desire to actually build things, things started to sort themselves out rather quickly.
Not saying that's what is going on here, but maybe it is helpful.
Maybe we can just not give it access to production databases ever?
Not picking on you, but AI maximalism has infected tech to the point where we talk about how to stop AI from deleting prod instead of seeing that giving AI access to prod is a foolish idea to begin with.
The most exasperating thing about the incident is how much of the media either tried to pin it on AI and/or Railway. The whole thing only took place because the guy FAFO’d by having AI work with prod directly.
Yet the narrative was mostly not about accountability for him. If I was a dumbass and deleted prod and wrote a post about it, nobody would care. Put an AI in there and all of the sudden it’s newsworthy. Ridiculous.
The takeaway here is to make this sort of scenario impossible in the future. It’s not hard to make that happen, but it might mean you need to manually interact with prod.