59 The untold secret: How to tame the Agent

59 The untold secret: How to tame the Agent I have written a very important 31 core analysis of AI value before, which mentioned the three major values of AI. Do you remember what they are? The most useful one is the third point: infinite nesting dolls - it is by far the most NB feature of LLM, bar none. This article will use actual cases to illustrate the use of my thinking and summary.

Recently, I often watch Lao Zeng on station B. I can learn practical routines in just a few minutes. I like it very much. I just happened to watch an episode of review today. He said that the term review comes from the game of Go. After each game, you think about it step by step from the beginning and re-take every step in your mind. This is called review. A standard review is divided into several steps. The first step is to objectively sort out what happened according to the timeline. The second step is to analyze in detail what was done well and what was not done well. What is the reason for winning and where is the loss? The third step is to improve after understanding. If you do it again, you should...