even if llm models stop getting smarter, there is so much room for innovation in the tools around them
the thing i’ve missed lately is the ability to dive in and start a new conversation within an existing one
rn its hard to set up many of them and next to impossible to keep track of
cannot believe I am actually sitting over here using cursor chat to generate instructions for cursor composer ... its feels like I keep hacking things to make it able to work with my project regardless of complexity
cursor composer in yolo mode is straight up like having two devs working with you all day
hopefully they start allowing us to bypass the max 25 tool calls per iteration
every morning i’m out driving around really early when there’s no one else on the road
only on christmas morning do the drunk people slowly run red lights and / or stop in the middle of the intersection
corporate rto is one of two types of executive flailing
1. “oh no 😓 i need to make investors feel like we’re going to be more productive”
2. “oh no 😓 i need to make investors feel like we’re going to cut costs, so im going to “fire” workers by forcing them to do something they can’t”
either way, count me out - companies where executives lead without being straight forward and honest to their employees and investors are on a downward spiral to inevitably being taken over by competition
all i really need an ai code editor to do is tell me me “yo, what in the heeeelllllll does this have to do with what you’re supposed to be working on?”