Slowing Down in the Age of Coding Agents

larve 15 points 2 comments March 24, 2026
the.scapegoat.dev · View on Hacker News

Discussion Highlights (2 comments)

asdfqwertzxcv

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?

2001zhaozhao

> 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.

Semantic search powered by Rivestack pgvector
3,471 stories · 32,344 chunks indexed