HNGlance

Top Best New Show

Adding verification steps to AI agents made them worse. I tested it 29 times

https://substack.com/@krzysztofdudek/p-192580258

by chrisdudek • 4 months ago

3 2 4 months ago

SU

Related Stories

Loading related stories...

Comments

2

Bjartr• 4 months ago

What works... External mechanical checks

This is what I expected verification to mean going in.

For some reason verification in this article means "try to convince the AI even harder than before"

chrisdudek• 4 months ago

I first tried to make the agent follow concrete paths, but no tricks worked. The reason I wanted to achieve it this way is that I wanted to create universal "rules" that would be followed without additional CLI or hooks. It didn't work out as expected. I needed it for my project, and currently I started to enforce more things with CLI, which is directly accessible by an agent, so it's now "guided" more forcefully.

Source preview

substack.com

HNGlance

© 2024-2026 HN Glance. Story content belongs to its original sources.

Privacy
Terms