←back to thread

Generative AI coding tools and agents do not work for me

(blog.miguelgrinberg.com)
399 points nomdep | 5 comments | 17 Jun 25 00:33 UTC | HN request time: 0.704s | source
1. p1dda ◴[17 Jun 25 02:49 UTC] No.44295314[source]▶
It would be interesting to see which is faster/better in competitive coding, the human coder or the human using AI to assist in coding.
replies(3): >>44295500 #>>44296033 #>>44296036 #
2. asciimov ◴[17 Jun 25 03:31 UTC] No.44295500[source]▶
It would only be interesting if the problem was truly novel. If the AI has already been trained on the problem it’ll just push out a solution.
3. wiseowise ◴[17 Jun 25 05:37 UTC] No.44296033[source]▶
It already happened. Last year AI submissions completely destroyed AoC, as far as I remember.
4. Snuggly73 ◴[17 Jun 25 05:38 UTC] No.44296036[source]▶
New benchmark for competitive coding dropped yesterday - https://livecodebenchpro.com/

Apparently models are not doing great for problems out of distribution.

replies(1): >>44296716 #
5. p1dda ◴[17 Jun 25 07:59 UTC] No.44296716[source]▶
It goes to show that the LLMs aren't intelligent in the way humans are. LLMs are a really great replacement for googling though