Post

P(aul) Frazee
@pfrazee.com

My impression is that AI coding is on a "pick two of three" triangle: scope, ship speed, and correctness. Small tools are (speed + correctness) fast prototypes are (speed + scope) and big projects are (scope + correctness). Every project that tries to be all three degrades to "fast prototype."

James Felix Black
@tft.io

It hits differently, and I would argue it hits differently based on experience (seniority is a thin proxy for this). The people I know most productive with the agents are the most senior folks.

P(aul) Frazee
@pfrazee.com

Yeah I'm curious whether that's a permanent relationship (experience -> productivity) or if experienced engineers are just picking the tools up faster

0 reposts10 likes

Steve
@steveax.bsky.social

I think it's a "know which areas the tricky/non-obvious bugs are likely hiding out in" sense that comes from experience that leads to prompting/specifying/pushing back on that makes for a good chunk of that

James Felix Black
@tft.io

I’m curious too because if it’s permanent (as I suspect, because I suspect the missing piece here is judgement), that’s very bad, because how do you train up juniors? If it’s the latter, that’s a lot less worrisome.

Kiwicmc
@kiwicmc.bsky.social

Us ‘senior’ folk trade AI replies and shake heads - “You are absolutely right to catch that! My previous code used a standard IF check, which is fine for most cases but technically has a tiny ‘race condition’ window.”