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

Generative AI coding tools and agents do not work for me

blog.miguelgrinberg.com

Post reply on HN
Print view View original

91–100 of 464 posts

Re: Generative AI coding tools and agents do not work for me

#91

Post by viraptor »

roxolotl wrote: post #22

> But interns learn and get better over time. The time that you spend reviewing code or providing feedback to an intern is not wasted, it is an investment in the future. The intern absorbs the knowledge you share and uses it for new tasks you assign to them later on. This is the piece that confuses me about the comparison to a junior or an intern. Humans learn about the business, the code, the history of the system.…

You don't have to reexplain the business context. Save it to the mdc file if it's important. The added benefit is that the next real person looking at the code can also use that to learn - it's actually cool for having good up to date documentation is now an asset.

Re: Generative AI coding tools and agents do not work for me

#92

Post by Terr_ »

xarope wrote: post #53
roxolotl wrote: post #22

> But interns learn and get better over time. The time that you spend reviewing code or providing feedback to an intern is not wasted, it is an investment in the future. The intern absorbs the knowledge you share and uses it for new tasks you assign to them later on. This is the piece that confuses me about the comparison to a junior or an intern. Humans learn about the business, the code, the history of the system.…

I think this is how certain LLMs end up with 14k worth of system prompts

"Be fast", "Be Cheap", "Be Good".

*dusts off hands* Problem solved! Man, am I great at management or what?

Re: Generative AI coding tools and agents do not work for me

#93

Post by 15123123 »

throwawaysleep wrote: post #81
asciimov wrote: post #67

Earlier quoted context omitted.

How are those without that kind of scratch supposed to keep up with those that do?

If you lack "that kind of scratch", you are at the learning stage for software development, not the keeping up stage. Either that or horribly underpaid.

$100 per month for a SaaS is quite a lot outside of Western countries. People are not even spending that much on VPN or Password Manager.

Re: Generative AI coding tools and agents do not work for me

#94

Post by bigbuppo »

ken47 wrote: post #10
frankc wrote: post #8

I just don't agree with this. I am generally telling the model how to do the work according to an architecture I specify using technology I understand. The hardest part for me in reviewing someone else's code is understanding their overall solution and how everything fits together as it's not likely to be exactly the way I would have structured the code or solved the problem. However, with an LLM it generally isn't s…

You can pre-agree on a solution path with human engineers too, with a similar effect.

Don't try to argue with those using AI coding tools. They don't interact well with actual humans, which is why they've been relegated to talking to the computer. We'll eventually have them all working on some busy projects to help with "marketing" to keep them distracted while the decent programmers that can actually work in a team environment can get back to useful work free of the terrible programmers and marketing departments.

Re: Generative AI coding tools and agents do not work for me

#95

Post by viraptor »

sothatsit wrote: post #89
viraptor wrote: post #78

Earlier quoted context omitted.

> It's something you can pick in a few minutes You can start in a few minutes, sure. (Also you can start using gdb in minutes) But GP is talking about the ceiling. Do you know which models work better for what kind of task? Do you know what format is better for extra files? Do you know when it's beneficial to restart / compress context? Are you using single prompts or multi stage planning trees? How are you managing…

I feel like there is also a very high ceiling to how much scaffolding you can produce for the agents to get them to work better. This includes custom prompts, custom CLAUDE.md files, other documentation files for Claude to read, and especially how well and quickly your linting and tests can run, and how much functionality they cover. That's not to mention MCP and getting Claude to talk to your database or open your w…

> then it will produce a specification document from that

I like a similar workflow where I iterate on the spec, then convert that into a plan, then feed that step by step to the agent, forcing full feature testing after each one.

Re: Generative AI coding tools and agents do not work for me

#96

Post by euleriancon »

abenga wrote: post #90
euleriancon wrote: post #71

> The truth that may be shocking to some is that open source contributions submitted by users do not really save me time either, because I also feel I have to do a rigorous review of them. This truly is shocking. If you are reviewing every single line of every package you intend to use how do you ever write any code?

You do not need to review every line of every package you use, just the subset of the interface you import/link and use. You have to review every line of code you commit into your project. I think attempting to equate the two is dishonest dissembling.

To me, the point the friend is making is, just like you said, that you don't need to review every line of code in a package, just the interface. The author misses the point that there truly is code that you trust without seeing it. At the moment AI code isn't as trustworthy as a well tested package but that isn't intrinsic to the technology, just a byproduct of the current state. As AI code becomes more reliable, it will likely become the case that you only need to read the subset of the interface you import/link and use.

Re: Generative AI coding tools and agents do not work for me

#97

Post by Terr_ »

hooverd wrote: post #44

The moat is that juniors, never having worked without these tools, provide revenue to AI middlemen. Ideally they're blasting their focus to hell on short form video and stimulants, and are mentally unable to do the job without them.

Given some the creeping appeal of LLMs as cheating tools in education, some of them may be arriving in the labor market with their brains already cooked.

Re: Generative AI coding tools and agents do not work for me

#98

Post by Winsaucerer »

dvt wrote: post #64

I'm actually quite bearish on AI in the generative space, but even I have to admit that writing boilerplate is "N" times faster using AI (use your favorite N). I hate when people claim this without any proof, so literally today this is what I asked ChatGPT: write a stub for a react context based on this section (which will function as a modal): ``` // a bunch of stuff ``` Worked great, it created a few files (the hoo…

This kind of thing is my main use, boilerplate stuff And for scripts that I don't care about -- e.g., if I need a quick bash script to do a once off task.

For harder problems, my experience is that it falls over, although I haven't been refining my LLM skills as much as some do. It seems that the bigger the project, the more it integrates with other things, the worse AI is. And moreover, for those tasks it's important for me or a human to do it because (a) we think about edge cases while we work through the problem intellectually, and (b) it gives us a deep understanding of the system.

Re: Generative AI coding tools and agents do not work for me

#99

Post by Jonovono »

marssaxman wrote: post #86

So far as I can tell, generative AI coding tools make the easy part of the job go faster, without helping with the hard part of the job - in fact, possibly making it harder. Coding just doesn't take that much time, and I don't need help doing it. You could make my coding output 100x faster without materially changing my overall productivity, so I simply don't bother to optimize there.

Are you a plumber perhaps?

Re: Generative AI coding tools and agents do not work for me

#100

Post by susshshshah »

sagarpatil wrote: post #87
jumploops wrote: post #2

> It takes me at least the same amount of time to review code not written by me than it would take me to write the code myself, if not more. As someone who uses Claude Code heavily, this is spot on. LLMs are great, but I find the more I cede control to them, the longer it takes to actually ship the code. I’ve found that the main benefit for me so far is the reduction of RSI symptoms, whereas the actual time savings a…

I always use Claude Code to debug issues, there’s no point in trying to do this yourself when AI can fix it in minutes (easy to verify if you write tests first) o3 with new search can do things in 5 mins that will take me at least 30 mins if I’m very efficient. Say what you want but the time savings is real.

How do you know what tests to write if you don’t understand the code?
Post reply on HN