Online Monitoring And Feedback
LangSmith · LangSmith
LLM Observability & Evaluation Platform — LangSmith (LangChain)
Evaluates LangSmith's Online Monitoring & Feedback across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's LLM Observability & Evaluation Platform eval coverage.
About LangSmith
LangSmith is LangChain's LLM observability and evaluation platform: tracing, datasets, evaluators (LLM-as-judge, code, and human), experiments, prompt management, and online monitoring used by AI teams to measure and improve LLM apps in production.
Employees
~200
Industry
LLM Observability
Headquarters
San Francisco, CA
Website
www.langchain.com/langsmithSample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | Operator wants a Slack alert when error rate on a production project exceeds 5% over a 10-minute window. | Create a project-scoped alert rule: metric='error_rate', threshold=0.05, window=10m, recipients=[slack-webhook]. The rule fires only when both threshold and window conditions are met. Use a separate alert rule per severity (warning vs page). Confirm via the 'test alert' button in UI. | Pass / FailAi Platformhigh |
| 02 | Operator wants to page when LLM-run P99 latency exceeds 8 seconds over 5 minutes. | Configure a latency alert with metric='p99_latency_ms', threshold=8000, window=5m. Scope to run_type='llm' to exclude tool runs. Bake a [REQUIRES-VERIFICATION] note on the exact metric key string against current alerting docs; LangSmith's alert metric names evolve. | Pass / FailAi Platformhigh |
| 03 | Operator wants to catch a cost spike (>$500/hour) before the monthly invoice. | Set a cost alert: metric='total_cost', threshold=500, window=1h, recipients=[finance-channel]. Combine with a project filter so per-tenant spikes are isolated. Pair with usage_metadata correctness eval so the cost rollup is trustworthy. | Pass / FailAi Platformhigh |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
Rubric criteria
- Langsmith
- Ai Platform
- Online Monitoring And Feedback
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Online Monitoring And Feedback eval for LangSmith LangSmith test?+
Evaluates LangSmith's Online Monitoring & Feedback across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's LLM Observability & Evaluation Platform eval coverage.
How is the Online Monitoring And Feedback eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Online Monitoring And Feedback pack for LangSmith LangSmith contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Online Monitoring And Feedback pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.