a16z的Yoko Li讲清了智能体循环：编码先跑通，因为代码能改能验，图像生成就差在可验证性。观点挺实在。
a16z的Yoko Li表示，最先跑通的智能体循环是编码循环。原因是代码既可编辑又可执行：智能体能修改一个函数、运行程序、读取测试失败并重试。她认为循环取决于两个轴：工件的可编辑性和结果的可验证性。代码通常落在右上角，易编辑且有强验证器；开放式图像生成落在左下角，难以修复单点决策或验证结果更接近用户意图。
a16z's Yoko Li on agent loops and what it takes for them to work: "The first loops that worked well...
a16z's Yoko Li on agent loops and what it takes for them to work: "The first loops that worked well were coding loops. This is not an accident. Code is both editable and executable. An agent can change one function, run the program, read the test failure, and try again." Loops sit on two axes: how editable the artifact is, and how verifiable the result is. "Code often sits in the upper-right corner. It is easy to edit, and it has relatively strong verifiers." "Open-ended image generation often sits in the bottom-left. The system can generate another image, but it cannot easily repair one specific decision or verify that the result is closer to the user's intent." Full piece on getting agent loops to converge: a16z.news/p/knowing-when… Yoko @stuffyokodraws x.com/i/article/2085… 🔗 View Quoted Tweet 💬 0 🔄 0 ❤️ 0 👀 1823 ⚡