跨国深谈← 返回←Andrew Garvin

① 他说过的话

6 条

我们的理念是提供更加详细、更清晰的错误信息,让智能体能够自我纠正。
Our perspective is to have much more verbose and clear errors so that the agent can self-correct.
—— Andrew Garvin · [06:53]

指向原始笔记的链接

从产品开发的角度看,我们的目标并不是让客户在没有人在环的情况下操作整个系统。
The goal that we have from a product development standpoint is not to have a customer operate the entire system without a human in the loop.
—— Andrew Garvin · [07:18]

指向原始笔记的链接

如果智能体可以成为产品并烧出高额的 token 账单,那么你能够对此进行计量就很重要。
If the agent can be the product and run up a token bill, it’s important for you to be able to meter on that.
—— Andrew Garvin · [10:00]

指向原始笔记的链接

再说一次,在那个世界里,拥有基于用量的定价模式很重要,因为有可能所有的价值基本上都归属于你平台的一个用户,在这个案例中就是一个智能体。
And again, in that world, it’s important for you to have a usage-based pricing model because you have the possibility of essentially all of the value accruing to essentially one user of your platform, which in this case would be an agent.
—— Andrew Garvin · [11:40]

指向原始笔记的链接

所有企业端编程智能体的情况——想想 Cognition 或 Cursor 或 OpenAI 和 Anthropic 自己——是他们开始采用更多承诺消费结构,就像云服务商们过去 10 年所做的那样。
What’s happening with all the coding agents in the enterprise, think like Cognition or Cursor or OpenAI and Anthropic themselves, is that they are starting to adopt more commit structures like the CSPs have done for the past 10 years.
—— Andrew Garvin · [12:23]

指向原始笔记的链接

再说一次,我们指导智能体构建这个的方式,只是用自然语言描述来复刻 Lovable 的定价模式。没有比这更难的东西了。
And again, the way that we coached the agent to be able to build this was just describing a natural language to replicate lovable’s pricing model. It was nothing more difficult than that.
—— Andrew Garvin · [15:41]

指向原始笔记的链接

② 出现在这些集

1 集

• 《智能体怎么付钱、怎么花钱:Metronome 讲透按量计费与智能体商务》 — 作为嘉宾

③ 他谈到的

点进去有真内容 —— 本页主要出口

Metronome · Stripe · Stripe Projects · OpenAI · Anthropic · Lovable · HubSpot · 智能体 · 编码智能体 · vibe coding

④ 也在聊「智能体」的人