Skip to content
LangChain Blog·· 1 d agoAI score38

LangSmith Alerts: Monitor LLM App Failures in Real Time

Catch production failures early with LangSmith Alerts

AI summary

LangChain launched Alerts in LangSmith, letting teams monitor LLM apps and agents in real time on error count and rate, average latency, and average feedback score. Each alert supports filters for specific run subsets, 5- or 15-minute aggregation windows, and thresholds, with delivery via PagerDuty or custom webhooks such as Slack. Users already sending production traces to LangSmith can set up an alert now.

Source:LangChain Blog · langchain.com