Skip to main content
LangSmith Engine is the LangSmith Agent for agent engineering. It works from your production traces to surface recurring failures, diagnose their root cause, and drive the fix across every stage of the development lifecycle. Each issue moves through a closed loop: a recurring failure is detected in your traces, the root cause is diagnosed, a fix is proposed, an evaluator is deployed to catch regressions, and if the issue resurfaces after being closed, Engine reopens it automatically.

​
Engine across the lifecycle

For each issue, Engine surfaces the contributing traces, proposes a fix, generates a custom evaluator to prevent regressions, and creates ground truth dataset examples from the production trace inputs.

Build: Open a pull request

Apply the proposed fix by opening a pull request in your connected repository. Engine can propose code changes to agents built with Deep Agents, LangChain, and LangGraph.

Test: Generate evaluators and datasets

Deploy a custom evaluator to catch regressions, and create ground truth dataset examples from production traces for offline evaluation.

Monitor: Detect recurring failures

Scan your tracing projects on a schedule to surface, prioritize, and diagnose recurring issues.

​
How Engine runs

Engine scans each connected tracing project every 6 hours, clustering and prioritizing issues by severity. It uses LangChain-managed inference and charges in LangChain Compute Units (LCUs). For setup, costs, and the full issue workflow, see Find and fix your agent’s failures.

​
Get started

Set up Engine

Enable Engine for your organization and configure it for a tracing project.

Engine webhook events

Forward detected issues into your incident-management, paging, or chat tools.
Connect these docs to Claude, VSCode, and more via MCP for real-time answers.
Edit this page on GitHub or file an issue.