
quite possibly this is just the early-days halo effect, it always takes a while to first start really noticing the ways a new model fucks up. also quite possibly i'm just not such a high percentile software engineer. but... this feels different

quite possibly this is just the early-days halo effect, it always takes a while to first start really noticing the ways a new model fucks up. also quite possibly i'm just not such a high percentile software engineer. but... this feels different

the previous shift was large but it still was mostly composed of things i was already doing and knew how to do well. maybe more of a product focus than before. but this is starting to feel like i need to rethink my job a little more fundamentally

if opus 4.5 started shifting us up the managerial stack so we wrote design docs instead of code, fable is definitely at least the start of the shift to the point where we're not really writing the design docs either

maybe this is what the ant people are talking about when they say "writing loops". it's impractical for ppl without their token spend budget, but if the models need this much less piloting then... if i'm not building robust outer loops what value am i actually adding?

ofc i knew the direction we were heading in. and i'm not at all worried about figuring out the right ways of interfacing here and continuing to climb the cyborgism management ladder. still it's a very strange moment.

@tenobrus It’s pretty great but it’s still prone to the “let’s create additional representations for stuff that already has some, write a thousand converters and also erase types and make everything stringly-typed just to ‘move faster with the prototype’”