LangSmith 推出 AI Search:SmithDB 把 traces 转成可读轨迹来搜索
LangChain 拆解了 LangSmith 新的 AI Search 内部机制:先把 trace 变成可读轨迹,再让模型逐条比对并流式返回,做 agent 调试的可以看看。
LangChain 介绍了 LangSmith 中 AI Search 的实现方式。SmithDB 先把 traces 转成可读的轨迹(trajectories),搜索时用结构化过滤器缩小候选范围。再由一个模型将剩余的每条轨迹与用户描述逐条比对,并流式返回匹配结果。
🧵 Behind the scenes of AI Search in LangSmith: TL;DR 1️⃣ SmithDB turns traces into readable trajectories. Structured filters narrow the candidates when you search 2️⃣ A model checks each remaining trajectory against your description and streams back matches Your browser does not support the video tag. 🔗 View on Twitter 💬 4 🔄 0 ❤️ 12 👀 1595 📊 4 ⚡