Hacker News Reader

Why not just split the tasks, and have different projects only house a portion of the code?

Why dumb down your work worse because the machine can't understand it?

Seems like the code is designed poorly if it can't be worked on in small chunks

Seems like you're dismissing the cost of artificially splitting what should be a monolith.

It outweighs the supposed productivity boost of LLMs by at least one order of magnitude if not more.

Most people have a smaller context window than LLMs.