跳到正文
Rohan Paul· @rohanpaul_ai · X·· 3 小时前AI 评分66

OpenAI 用 Codex 智能体自动排查 ChatGPT 性能问题,诊断准确但修复落地率低

AI 导读

Rohan Paul 转述 OpenAI 的 Performance Factory 系统:ChatGPT 变慢的告警会自动启动 Codex 智能体翻查日志和近期代码变更定位原因,第二个智能体尝试推翻该诊断,再由另一个提出修复方案并交人工工程师批准。该系统每天约进行 17 次调查,其中约 70% 被判定准确;但修复是短板,600 多例确诊问题中只有约 15 个智能体自己的修复原样上线。

正文 · 原文

OpenAI built a system where alerts about ChatGPT slowing down automatically start Codex agents, which dig through logs and recent code changes to find the cause.
A second agent then tries to prove that diagnosis wrong, and another proposes a code fix that a human engineer approves.
It finds causes well, about 17 investigations a day with 70% judged accurate, but fixing is the weak part: out of 600-plus diagnosed problems, only about 15 of the agents’ own fixes shipped unchanged.

来源:Rohan Paul · x.com

事件后续 · 1

查看事件全部