Posts
All the articles I've posted.
大模型结果如何测试和验证:Anthropic 与 OpenAI 的方法论
Published: at 02:02 AMEval 不是给模型出题,而是为一个 Claim 构造可重复实验:判定、效度、评分器、指标、生产飞轮
Dev weekly 2026-Sep-07
Published: at 02:02 AMDev weekly
Dev weekly 2026-Aug-31
Published: at 02:02 AMDev weekly
Dev weekly 2026-Aug-24
Published: at 02:02 AMDev weekly
AI Agent interview questions
Published: at 02:02 AM2026 年 AI Agent 工程师面试高频题与资料汇总,按主题归类并附真实来源