Comment code writing (Score 1) 172
our team has broadly switched over for 90% of all code writing. all small changes, all minor updates, and all exploratory work is done via AI 100%. When we get to production deployed code, its up and down (mostly up) that AI can write solid, working code, but the more complex the task, the more important measure is who prompts the AI.
Detailed, longwinded prompts that really restrict and direct the coding task precisely seems to allow nearly no human coding to be done. But that requires our best software architects to sit down for a couple hours and really work out a precise prompt, which involves having the knowledge and skills to not just code the project end to end, but also to precisely determine what you want coded at the outset. no throwing pasta at the wall to see if it sticks.
Our junior folks struggle with this because they don't normally have a clear vision of exactly what they want coded and the final architecture until they are halfway through the task. we are obviously seeing this with efficiency as well (tokens/task completed). But they are learning quite a bit quicker.