Anthropic 工程师讲 harness 与自改进 agent

fomoebka @fomoebka原推发布于 2026.10.24

在 X 上打开原推

This is f*cking insane. Anthropic engineer: "If you want to be in the top 1% of AI engineers, stop writing prompts. Build a harness that improves itself." In 41 minutes, she explains how Anthropic builds with AI internally and why harnesses and graphs sit at the center of it Prompts → Agents → Harness → Loops → Graphs A harness makes each unit of work correct without you A graph decides which units should exist and turns every accepted result into a rule for the next run That is what "self-improving" actually means. Almost nobody builds it Worth more than most $500 agent engineering courses Bookmark and watch it today Then start with the "Your First Grok Bot" guide below and give one agent a real job ↓

这条是引用着另一条写的：x.com/m4lkxx/status/2111895215556456480