How much can you delegate to agents? - by Jina Yoon

kept by eddie jul 28

The article presents a practical framework for deciding how much autonomy to grant AI agents based on task checkability and undo cost, outlining four levels from assistant to self‑driving mode.

Why we're bullish on loops - by Ian Vanagas
Loops—self-prompting agents that autonomously complete long-running tasks using context and verification—are becoming viable due to model improvements and enable self-driving products.
also on PostHog, #automation, #ai-agents, #ai-coding, #developer-tools
Autoresearch, Claude and Constrained Optimization
An experiment using Claude Code to autonomously optimize a file compression algorithm over 10 iterations, achieving competitive results against standard tools by iterating on an LZSS-based approach.
also on PostHog, #automation, #ai-coding, #developer-tools