Live data from Hacker News

Vibe Coding Is Overrated

transitions.substack.com

1–10 of 93 posts

Re: Vibe Coding Is Overrated

#3
Vibe coding is definitely problematic for someone who isn't a strong coder, aka me. Recently for fun, I used ChatGPT to code a python script to scrape another news site and render it in the style of HackerNews.

It took 21 iterations of code updates to get a working version. It isn't bad but doesn't pull the latest articles (I don't know why) and hell if I could debug it.

Re: Vibe Coding Is Overrated

#4
This is not news for people who already know how to code. The problem is that there are tens of thousands of people who have no idea what they're doing, yet they evangelize their hollow victories to anyone willing to listen.

Worse yet, there are grifters like levelsio who will pretend like it's great because they know that's what their followers want to hear.

Re: Vibe Coding Is Overrated

#5
post #3

Vibe coding is definitely problematic for someone who isn't a strong coder, aka me. Recently for fun, I used ChatGPT to code a python script to scrape another news site and render it in the style of HackerNews. It took 21 iterations of code updates to get a working version. It isn't bad but doesn't pull the latest articles (I don't know why) and hell if I could debug it.

I've had a similar experience. It requires strong nerves and a lot of patience.

Re: Vibe Coding Is Overrated

#8
One thing implied or embedded in Andrej Karpathy’s description, that I think explains the difference between influencers experience vs lay people’s experience, and is not propagated in others description of vibe coding, is that one of the essential pieces of vibe coding is that you can’t care too much about the outcome. Like if you have a specific feature in mind, the LLM might not understand how to generate it, so you might just have to skip it.

Re: Vibe Coding Is Overrated

#9
After a week of trying to vibe cdk code, then sitting down in 4 hours to do the whole thing myself I can confirm. While it is now a good tool to combat fatigue, research and prototype, it still struggles with context switching, arid knowledge domains and long-feedback-loop systems
Post reply on HN