用 Jev 审计工作流、砍掉昂贵模型调用

Gipp 🦅 @gippp69原推发布于 2026.09.28

在 X 上打开原推

Send this Jev prompt to any LLM or AI agent it audits your workflow, finds where you’re burning tokens, time and money, then moves repeatable decisions out of expensive model calls in the desk example, 20,000 Jev decisions cost just $0.84. only 4% escalated to writing, but those 800 drafts added another $24 that’s the real alpha: don’t optimize the expensive model first. reduce how often you need to call it at all Jev handles the cheap picks, Grok Bot writes only when needed, and code keeps control of the risky stuff

这条是引用着另一条写的：x.com/gippp69/status/2103394701063737504