企业需要让数据、人员和 Agent 协同的方案，Glass 的思路值得关注
精选理由
作者聊了个观察：各公司都在找能把数据和 Agent 串起来的工具，Glass 思路不错但还没实测，适合想引入 Agent 协作的人参考。
svpino 在推文中观察到，他接触的公司都在做或想买把数据、人员和 Agent 拉到同一页面上的工具。他引用文章的观点：只给员工最贵的乐器奏不出交响乐，光靠 tokens 和模型不够。他认为关键在于让模型从历史经验中学习并自动改进。他未实测 Glass，只对其思路表示认同。
原文 · Ate-a-Pi
Every company I know is either building or looking to buy something that puts their data, people, and agents on the same page.
I haven't tried Glass, so I can't speak about how it works, but the idea is the right one:
You won't get a symphony by giving your employees the most expensive instruments. Tokens and models alone are not enough. (I'm paraphrasing the article here.)
We need effective ways for our models to learn from the past and automatically improve.
