Hacker News new | ask | show | jobs
by hsn915 267 days ago
It is time to acknowledge that AI coding does not actually work.

ok, you think it's a promising field and you want to explore it, fine. Go for it.

Just stop pretending that what these models are currently doing is good enough to replace programmers.

I use LLMs a lot, even for explaining documentation.

I used to use them for writing _some_ code, but I have never ever gotten a code sample over 10 lines that was not in need of heavy modifications to make it work correctly.

Some people are pretending to write hundreds of lines of code with LLMs, even entire applications. All I have to say is "lol".

5 comments

I have personally witnessed Cursor diagnose and then fix an actual non-trivial bug starting from its description in Jira, and I only had to course-correct it once.

I have also seen it fail on far simpler tasks.

It varies so much depending on what you are doing, the language etc that generic proclamations "it works!" or "it doesn't work!" are pretty much meaningless.

That aside, you seem to be conflating "it works" with "good enough to replace programmers", but these aren't synonyms.

And on the gripping hand, one way to "make" it work is simply to lower the standards. Which our industry has been doing aplenty for a long time now even before AI, so we shouldn't be surprised when top management drives it to its logical completion.

Very interesting observation.

I haven’t written a function by hand in 18 months.

Same. I haven't written any code by hand in some time. Oh well. I guess I'm just doing it wrong.
Have you built anything public that folks can try out?

Not doubting but it helps to contextualize things

https://app.grantpuma.com/

It's a startup for finding grants. We have california state, federal, non-profit and california city/county grants. My landing page absolutely sucks but if you sign up / upload some papers or make some search cards you'll like the experience.

I'm very excited to try out the new Qwen XL that came out recently for visual design. I could really use some better communication to users of the capabilities of the platform.

Landing site is completely broken on Safari, iOS 26, stable channel, iPhone 17 Air

A lot of content is off-screen, and can't be swiped back on-screen.

-

Actual app did let me theough the flow, but had things like dialogs staying open and covering content (I assume a non tech person would have missed them)

I think part of the AI gap right now is the UI/UX you're expecting. AI isn't able to meet the bar for UI work for me at all right now. I do use it via Tab completion, or chat for specific scaffolding, but the agentic form is rough

There are so many people en masse who simply cannot accept that the days of coding manually, by hand, could be coming to an end. It's wild.
Have you built anything in 18 months? I keep asking to see these apps that people supposedly vibe coded in a weekend but when I ask them to share it, nothing.
https://www.gabrieluribe.me

I have some publicly accessible projects there.

Ah, a $999 vibe coding course. That completely checks out.
That is one of many things! I offer that because I've had enough people request it over the past year.

Otherwise, I do professional software consulting + a variety of projects as an independent.

Can you show me a sample of the code you have AI write for you?
I have 20 years of experience and am now just running Opus 4.1 all the time. Of course it’s possible.

In my case I found having it always Ultrathink and to always work with TDD to work well. Also you have to use Plan Mode first and refine the plan. “What clarifying questions do you have for me” prompts me with a handful of numbered questions that are always really beneficial for refining the plan.

every attempt I've attempted thus far has failed at first run, but it was close to a solution that could be adapted and fixed. This has been especially helpful in areas where I'm charting very unfamiliar territory for very narrow scoped problems.

But hearing your 10 line constraint gives me a very https://en.wikipedia.org/wiki/Infinite_monkey_theorem vibe to the challenge.

sounds like poor prompt engineering. Devin and Claude can both do better work than many interns I have mentored in my career, and faster too. We likely have many many years until it will be even close to replacing an experienced developer but we are already at the point it IS replacing junior engineers.

Whether you agree or not, the market has spoken. New grad hiring is WAY down. Fresh CS grads are having an hell of a time finding work compared to 2 years ago.

Got proof that hiring is down due to AI?