Comment by sho_hn
6 days ago
I doubt this holds true generally. The smart coders I know who are also LLM users generally develop a decent intuition for what the models are good and bad at, and how to steer them into good performance.
6 days ago
I doubt this holds true generally. The smart coders I know who are also LLM users generally develop a decent intuition for what the models are good and bad at, and how to steer them into good performance.
Then perhaps my friend has remained a skeptic for so long that he's atrophied in this regard (which OP's post touches on). Either way, most of his day job is as a CTO/manager at a startup, so he's not in the weeds coding as much anymore in the first place. I should've seen how he prompts LLMs for managerial tasks, then I'd know whether his 'prompt laziness' was systemic or influenced by his coding knowledge.