Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A coder considers the waning days of the craft

newyorker.com

Post reply on HN
Print view View original

801–810 of 1001 posts

Re: A coder considers the waning days of the craft

#801

Post by coffeebeqn »

happytiger wrote: post #441
ikari_pl wrote: post #421

Earlier quoted context omitted.

4th option. They're trying to enter the market and get a good job, but some companies consider them less experienced but more expensive than Copilot.

Fewer but more senior resources seems possible. I have already seen the impact of AI systems, reducing the number of junior coding positions and outsourcing. On the other hand, anyone who actually takes the time to get good with AI coding tools can be performing at senior levels, much more quickly, so there’s that too. I’m not convinced that it’s some kind of apocalypse as much as it is a massive shift that is going…

I can kind of see that with contractors on the lower end fairly soon. If I put the same amount of effort into telling a programming specific GPT that understands requirements properly as I do with some contractors I don’t think that is a long way away from today.

For juniors - I want them to become eventually independent so I don’t have to worry about them and their tasks (and can give them vaguer tasks) so it’s a different goal.

Re: A coder considers the waning days of the craft

#803

Post by muldvarp »

alm1 wrote: post #775
RagnarD wrote: post #614

Not to be unkind but simply factual: the author is clearly not a great developer. e.g.: "At one point, we wanted a command that would print a hundred random lines from a dictionary file. I thought about the problem for a few minutes, and, when thinking failed, tried Googling." Really? "But he had no idea what a pain it is to make an iPhone app. I’d tried a few times and never got beyond something that half worked. I…

In author's defense he doesn't claim to be exceptional and even concludes his essay with labeling this A.I. moment as “the revenge of the so-so programmer.” I actually enjoyed author's nuanced take, different from the common narrative of AI takeover. To the contrary I think LLM-driven coding is not a tool of power users, but rather a tool of disempowered users. They will bring many more people into the engineering pr…

I don't think a "so-so programmer" would need to google to figure out how to select random lines from a file.

Re: A coder considers the waning days of the craft

#804

Post by blharr »

jameshart wrote: post #784
kypro wrote: post #727

Earlier quoted context omitted.

You may think it did a good job because of your limited CSS ability. I'd be amazed if ChatGPT can create pixel-perfect animations and transitions along with reusable clean CSS code which supports all of the browser requirements at your org. I've seen the similar claims made on Twitter by people with zero programming ability claiming they've used ChatGPT to build an app. Although 99% of the time what they've actually…

Here’s the thing though: If a new version of the app can be generated on the fly in minutes, why would we need to worry about reusability? GPT generated software can be disposable . Why even check the source code in to git - the original source artifact is the prompt after all.

Because in my experience GPT can produce a maximum of like 200 lines of code before it makes an error usually.

Re: A coder considers the waning days of the craft

#805

Post by manicennui »

dsQTbR7Y5mRHnZv wrote: post #346
manicennui wrote: post #228

Earlier quoted context omitted.

"AI" helps the people who rely on Stack Overflow to do their jobs. Then they come to HN and worry that we will all be out of jobs soon. The problem it solves is "I don't actually know what I am doing at all". These people are not a threat to anyone competent and no amount of questionable code generated by "AI" will change that.

What domain do you work in that you never reference stack overflow?

It's not that I never reference Stack Overflow; it is just not frequent. I actually know my tools fairly well and most corporate development work simply isn't challenging. I also use vim with nothing but syntax highlighting.

Re: A coder considers the waning days of the craft

#806

Post by whynotminot »

gessha wrote: post #788
steveBK123 wrote: post #646

Earlier quoted context omitted.

That's about my experience. The worst dev on my team uses ChatGPT a lot, and its facilitated him producing more bad code more quickly. I'm not sure it's a win for anyone, and he's still unlikely to be with the team in a year. It allows a dev who doesn't care about their craft or improving to generate code without learning anything. The code they generate today or a year from today is the same quality. Part of it is t…

I have a hunch that using ChatGPT might be a skill in of itself and it doesn’t necessarily hurt or help any particular skill level of developers. In previous replies in this thread the claim is it helps novices compete with associates or associates with seniors but in reality, it will probably help any tier of skill level. You just have to figure out how to prompt it

One hundred percent. Most people I’ve seen dismiss ChatGPT simply refuse to engage it appropriately. It’s not likely to solve your most complex problem with a single prompt.

Asking the right questions is such an important skill in and of itself. I think we’re seeing to some extent the old joke about engineers not knowing how to talk to people manifest itself a bit with a lot of engineers right now not knowing quite how to get good results from ChatGPT. Sort of looking around the room wondering what they’re missing since it seems quite dumb to them.

Re: A coder considers the waning days of the craft

#807

Post by electrondood »

Exuma wrote: post #640
anonzzzies wrote: post #535

Earlier quoted context omitted.

So, don't leave us in suspense; what do you ask of it? Because I'm quite sure it can already pass it. Your experience is very different from mine anyway. I am a grumpy old backend dev that uses formal verification in anger when I consider it is needed and who gets annoyed when things don't act logical. We are working with computers, so everything is logical, but no; I mean things like a lot of frontend stuff. I ask o…

How do you make a custom gpt which knows a specific code base? I have been wanting to do this

You tune an existing model on your own set of inputs/outputs.

Whatever you expect to start typing, and have the model produce as output, should be those input/output pairs.

I'd start by using ChatGPT etc. to add comments throughout your code base describing the code. Then break it into pairs where the input is the prefacing comment, and the output is the code that follows. Create about 400-500 such pairs, and train a model with 3-4 epochs.

Some concerns: you're going to get output that looks like your existing codebase, so if it's crap, you'll create a function which can produce crap from comments. :-)

Re: A coder considers the waning days of the craft

#808

Post by wkipling »

mike_hearn wrote: post #456
tangjurine wrote: post #402

Earlier quoted context omitted.

The author isn't a professional programmer. He has a couple side projects and doesn't seem very good at coding.

He is. Later in the article he explains a task he did during the working day to his wife. It seemed like his employer uses some sort of custom tableview control, or he was working on a business report. Honestly it sounded like the bigger threat to that particular task was someone finding an open source library that did it already, but it illustrated the point and the story was nicely written.

Yeah there's lot's of people in this thread who didn't read the full article and the conclusion the Author comes to.

Re: A coder considers the waning days of the craft

#809

Post by dsego »

civilitty wrote: post #759
dsego wrote: post #484

Earlier quoted context omitted.

Honestly, this reminds me of my early days when auto-complete and OOP was the best thing. It's why OOP proliferated, you type in the name of the object and get a list of completions in visual studio. No need to read docs or understand what you are doing, just find the method you need on that object. This new copilot coding reminds me of that, throwing shit and see what sticks, without understanding why and how it wor…

> I can't help to shake the thought that we will see heaps of mediocre code cranked out that just about works, even though the problem maybe required a tenth of that to solve it. That’s just more of the same, isn’t it? IMO that’s already the current state of the industry.

I think you're right.

Re: A coder considers the waning days of the craft

#810

Post by emporas »

wokwokwok wrote: post #564
throwaway346434 wrote: post #537

Earlier quoted context omitted.

... What. It hallucinates and it doesn't compile, fine. It hallucinates and flips a 1 with a -1; oops that's a lot of lost revenue. But it compiled, right? It hallucinates, and in 4% of cases rejects a home loan when it shouldn't because of a convoluted set of nested conditions, only there is no one on staff that can explain the logic of why something is laid out the way it is and I mean, it works 96% of the time so…

The output of an LLM is a distribution, and yes, if you’re just taking the first answer, that’s problematic. However, it is a distribution, and than means the majority of solutions are not weird edge cases, they’re valid solutions. Your job as a user is to generate multiple solutions and then review them and pick the one you like the most, and maybe modify it to work correctly if it has weird edge cases. How do you d…

In general, one should not instruct GPT to solve a problem. The instructions should be about generating code, after a human thought process took place, and then generate even more code, then even more, and after merging all the code together the problem is solved.

The particulars are roughly what you describe, in how to achieve that.

Post reply on HN