FRESH

Hacker News

Slowing Down in the Age of Coding Agents

15 points by larve

by 2001zhaozhao

0 subcomment

> I usually have one or two tasks running, but I don't feel pressure to maximize parallelism. Most of my brain is already occupied with thinking hard about the shape of what I'm building — and that work can't be parallelized. I can't do more; anything on top would encure a task switching cost.
Maybe there should be new agent UI's that help you get the most out of a singular work stream.
For example being able to spin up an asynchronous agent in a side window that can iterate with you on ideas while the main agent works, and then send any outputs to the main agent easily.

by asdfqwertzxcv

0 subcomment

I've been doing similar: analog-ish for deep thinking and its been akin to the 'measure twice, cut once' adage and I've noticed the projects have improved outcomes. PS. If OP sees this: How are you liking the Pentax 17?