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

Senior Developer Skills in the AI Age

manuel.kiessling.net

Post reply on HN
Print view View original

201–210 of 323 posts

Re: Senior Developer Skills in the AI Age

#201

Post by echelon »

dghlsakjg wrote: post #111
echelon wrote: post #78

Earlier quoted context omitted.

Photoshop is unapproachable to the 99%. A faster GPT 4o will kill Photoshop for good.

Photoshop is a tool designed for the 1% of people who want that level of control for their vision. Adobe has several other tools for other markets. Even the latest model from this week, which is undeniably impressive, can’t get close to the level of control that photoshop gives me. It often edits parts of the image I haven’t asked it to touch among other issues. I use photoshop as a former photojournalist, and AI man…

This rhymes with still developing your own film.

I'm really eager to see how this pans out in a decade.

Re: Senior Developer Skills in the AI Age

#202

Post by baq »

XorNot wrote: post #193
dheera wrote: post #158

Earlier quoted context omitted.

> You're not wrong here, but there's a big difference in programming one-off tooling or prototype MVPs and programming things that need to be maintained for years and years. Humans also worry about their jobs, especially in PIP-happy companies; they are very well known for writing intentionally over-complicated code that only they understand so that they are irreplaceable

I'm not convinced this actually happens. Seems more like somthing people assume happens because they don't like whatever codebase is at the new job.

If your TC is 500k-1M and you don’t feel like job hopping anymore, you’d certainly not want to get hit by a random layoff due to insufficient organizational masculinity or whatever. Maintaining a complex blob of mission critical code is one way of increasing your survival chances, though of course nothing is guaranteed.

Re: Senior Developer Skills in the AI Age

#203

Post by cztomsik »

Aurornis wrote: post #36
milicat wrote: post #24

Earlier quoted context omitted.

The more I browse through this, the more I agree. I feel like one could delete almost all comments from that project without losing any information – which means, at least the variable naming is (probably?) sensible. Then again, I don't know the application domain. Also… def _save_current_date_time(current_date_time_file: str, current_date_time: str) -> None: with Path(current_date_time_file).open("w") as f: f.write(…

> I feel like one could delete almost all comments from that project without losing any information I far from a heavy LLM coder but I’ve noticed a massive excess of unnecessary comments in most output. I’m always deleting the obvious ones. But then I started noticing that the comments seem to help the LLM navigate additional code changes. It’s like a big trail of breadcrumbs for the LLM to parse. I wouldn’t be surpr…

More tokens -> more compute involved. Attention-based models work by attending every token with each other, so more tokens means not only having more time to "think" but also being able to think "better". That is also at least part of the reason why o1/o3/R1 can sometimes solve what other LLMs could not.

Anyway, I don't think any of the current LLMs are really good for coding. What it's good at is copy-pasting (with some minor changes) from the massive code corpus it has been pre-trained. For example, give it some Zig code and it's straight unable to solve even basic tasks. Same if you give it really unique task, or if you simply ask for potential improvements of your existing code. Very, very bad results, no signs of out-of-box thinking whatsoever.

BTW: I think what people are missing is that LLMs are really great at language modeling. I had great results, and boosts in productivity, just by being able to prepare the task specification, and do quick changes in that really easily. Once I have a good understanding of the problem, I can usually implement everything quickly, and do it in much much better way than any LLM can currently do.

Re: Senior Developer Skills in the AI Age

#204

Post by FeepingCreature »

nottorp wrote: post #186
FeepingCreature wrote: post #172

Earlier quoted context omitted.

> What if I had trusted the code? It was working after all. Then you would have been done five minutes earlier? I mean, this sort of reads like a parody of microoptimization.

No, it reads like "your precious AI generates first year junior code". Like the original article.

There is nothing wrong with first year junior code that does the job.

Re: Senior Developer Skills in the AI Age

#205

Post by miningape »

nottorp wrote: post #184
bongodongobob wrote: post #131

Earlier quoted context omitted.

Why would you blindly trust any code? Did you tell it to optimize for speed? If not, why are you surprised it didn't?

So, most low level functions that enumerate the files in a directory return a structure that contains the file data from each file. Including size. You already have it in memory. Your brilliant AI calls another low level function to get the file size on the file name . (also did worse stuff but let's not go into details). Do you call reading the file size from the in memory structure that you already have a speed opt…

Yep exactly, LLMs blunder over the most simple nonsense and just leaves a mess in their wake. This isn't a mistake you could make if you actually understood what the library is doing / is returning.

It's so funny how these AI bros make excuse after excuse for glaring issues rather than just accept AI doesn't actually understand what it's doing (not even considering it's faster to just write good quality code on the first try).

Re: Senior Developer Skills in the AI Age

#206

Post by theteapot »

necovek wrote: post #9

The premise might possibly be true, but as an actually seasoned Python developer, I've taken a look at one file: https://github.com/dx-tooling/platform-problem-monitoring-co... All of it smells of a (lousy) junior software engineer: from configuring root logger at the top, module level (which relies on module import caching not to be reapplied), over not using a stdlib config file parser and building one themselves,…

> to a raciness in load_json where it's checked for file existence with an if and then carrying on as if the file is certainly there...

It's not a race. It's just redundant. If the file does not exist at the time you actually try to access it you get the same error with slightly better error message.

Re: Senior Developer Skills in the AI Age

#207

Post by LtWorf »

baq wrote: post #202
XorNot wrote: post #193

Earlier quoted context omitted.

I'm not convinced this actually happens. Seems more like somthing people assume happens because they don't like whatever codebase is at the new job.

If your TC is 500k-1M and you don’t feel like job hopping anymore, you’d certainly not want to get hit by a random layoff due to insufficient organizational masculinity or whatever. Maintaining a complex blob of mission critical code is one way of increasing your survival chances, though of course nothing is guaranteed.

People doing layoffs have no idea of who works and who's warming the chair.

Re: Senior Developer Skills in the AI Age

#208

Post by LtWorf »

ramesh31 wrote: post #76
nsonha wrote: post #67

Earlier quoted context omitted.

Why do you HAVE TO one-shot? No one says you have to code like those influencers. You are a software engineer, use AI like one, iteratively.

>No one says you have to code like those influencers. You are a software engineer, use AI like one, iteratively. This is my issue with all the AI naysayers at this point. It seems to all boil down to "haha, stupid noob can't code so he uses AI" in their minds. It's like they are incapable of understanding that there could simultaneously be a bunch of junior devs pushing greenfield YouTube demos of vibe coding, while…

Weren't you the guy who only writes HTML? Maybe let domain experts comment on their domain of expertise.

Re: Senior Developer Skills in the AI Age

#209

Post by carpo »

theshrike79 wrote: post #199
carpo wrote: post #63

Earlier quoted context omitted.

I'm the complete opposite. After being burnt out and feeling an almost physical repulsion to starting anything new, using AI has renewed my passion. I've almost finished a side project I started 4 weeks ago and it's been awesome. Used AI from the beginning for a Desktop app with a framework I'd never heard of before and the learning curve is almost non-existent. To be able to get the boring things done in minutes is…

I'm on this boat. I can use LLMs to skip the boring bits like generating API glue classes or simple output functions. Example: I'm building a tool to grab my data from different sites like Steam, Imdb, Letterboxd and Goodreads. I know perfectly well how to write a parser for the Goodreads CSV output, but it doesn't exactly tickle my brain. Cursor or Cline will do it in minutes. Now I've got some data to work with, wh…

Yeah, I've found it great at XML too. All sorts of boilerplate stuff works so well. As it the project gets bigger Ive had to think for longer about things and step in more, but as you said, that's usually the fun, meaty bit that you want to work on anyway. My app is about 15k lines of code now, and it's a bit more work than at the beginning.

Re: Senior Developer Skills in the AI Age

#210

Post by miningape »

skydhash wrote: post #51
rs186 wrote: post #19

That prompt looks horrifying. I am not going to spend half an hour coming up with that prompt, tweaking it, and then spend many hours (on the optimistic side) to track down all the hallucinated code and hidden bugs. Have been there once, never going to do that again. I'd rather do it myself to have a piece of mind.

I wonder how much time it would take with some samples from GitHub, and various documentation about python laying around (languages, cheatsheet, libraries)...

10 minutes tops, once you have the idea and you've thought it through a bit just spamming out the code isn't that hard nor does it take that long. If there's anything surprising (e.g. function call returns something a bit different than expected) it's fast enough to just read the docs and change your mental model slightly.
Post reply on HN