There’s this big debate erupting now and then if #LLM-s are groundbreaking technology for writing code.
They absolutely are. Writing code is easier, especially the boring part we would „automate” or script in the past. Now we don’t have to.
But… Writing code itself is the easiest part so not a great unlock.
Models like #GPT4, #Mistral #Large2 or #Sonnet35 also help with debugging, rubber-ducking and “querying” documentation. This is where the impact is most pronounced in my day to day work.
=> More informations about this toot | More toots from oktawian@mastodon.social
text/gemini
This content has been proxied by September (ba2dc).