LangSmith AI Search 升级:自然语言解析为结构化过滤器
LangSmith 这次更新挺实用:直接用一句人话就能查 traces,模糊提问它还会先追问。做 LLM 应用排查 bug、复盘用户反馈的工程师可以试试,比手写过滤器省事多了。
LangChain 对 LangSmith 的 AI Search 进行了更新。用户可以用自然语言提问,例如“本周因 LLM 报错而受挫的用户”,系统会将查询解析成结构化过滤器,并附加 AI 检查来覆盖过滤器无法表达的条件。
当问题表述模糊时,AI Search 会先提出追问再执行搜索,而不是直接返回不准确的结果。这一机制让 trace 检索从关键词匹配转向意图理解,开发者可以用更少的过滤语法完成同样的排查工作。
LangSmith AI Search got a whole lot smarter ✨ Ask for what you want in plain words, like "frustrated users with LLM errors this week," and AI search parses your query into structured filters, plus an AI check for anything a filter can't express. When your question is vague, it asks a follow-up before it searches. Try it out today ➡️ docs.langchain.com/langsmith/filt…
This report is categorized under AITOP's "Products" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include LangSmith, LangChain, observability, developer tools.
Source attribution: curated from LangChain (@LangChainAI).