LangChain Launches LangSmith AI Search for Semantic Agent Trace Retrieval
LangChain introduced AI Search in LangSmith, letting users retrieve agent traces by natural-language behavior semantics. SmithDB normalizes traces into a standard message format, and a small fast model checks candidates in parallel, each taking under half a second with streaming results.
2026-10-17 ~ 2026-10-17 · 4 related posts
- LangChain explains LangSmith AI Search: trajectory views plus a small parallel model under half a second per check — LangChain · 2026-10-17
- SmithDB converts raw traces into a standard message format and builds readable agent trajectory views — LangChain · 2026-10-17
- LangSmith search: parallel model checks under half a second, matches stream back before full scan — LangChain · 2026-10-17
- LangSmith AI Search: query agent traces by behavior, median Open SWE trace runs 84K tokens — LangChain · 2026-10-17