Get Started with Insights Agent in LangSmith

Today's popular agents produce millions of traces per day—soon to be billions. These traces contain valuable signal about an agent's capabilities and how real users engage with it. If you could review each interaction, you would gain deep insight into how to improve your agent. Manual review is time-consuming and impossible at scale, so how can we automate this insight generation process? Insights Agent is our first step towards helping LangSmith users find signal in their production traces. Insights Agent analyzes traces to discover and surface common usage patterns, agent behaviors, and failure modes. In agent engineering, you need to iterate rapidly to build reliable experiences. This new feature helps you figure out where to focus your next set of tests based on real interactions your agent is having. Learn more: https://bit.ly/4qlw15V Get started with LangSmith: https://bit.ly/43vKzGj Learn how to observe and evaluate agents with LangSmith on LangChain Academy: https://academy.langchain.com/courses/intro-to-langsmith/?utm_medium=social&utm_source=youtube&utm_campaign=q4-2025_youtube-academy-links_aw Observe, evaluate, and deploy agents with LangSmith: https://smith.langchain.com/?utm_medium=social&utm_source=youtube&utm_campaign=q4-2025_youtube-links_aw