Guides

End-to-end examples of how to use Langfuse: tutorials, runnable cookbooks, and video walkthroughs.

Introduction to Langfuse

If you are new to Langfuse, watch this walkthrough to get a thorough introduction to the Langfuse platform.

Evaluation Tutorials

Start with the fundamentals and progress through advanced techniques as your application complexity grows. Read our blog post for an overview.

Set Up a User Feedback Loop

Capture explicit and implicit user feedback as scores on your traces, and use it to surface the responses worth reviewing. Error Analysis

Review traces to classify issues like hallucinations, irrelevance, and formatting errors. Turn raw logs into actionable insights. Automated Evaluations

Automated evaluators give you a scalable way to measure and monitor your application’s failure modes. Testing LLM Applications

Build a testing foundation with deterministic checks and probabilistic ones using LLM judges to prevent regressions. CI/CD for Experiments

Run Langfuse experiments in CI/CD pipelines to catch quality regressions before they reach production. Synthetic Dataset Generation

Scale your test coverage with LLM-generated synthetic data. Create diverse inputs and adversarial cases without waiting for users. Evaluate and Improve Structured Extraction

Score document extraction per field with a labeled dataset, catch regressions that pooled accuracy hides, and iterate with experiments. Experiment Interpretation

Run experiments to compare prompts, models, or pipelines. Measure metrics like accuracy and speed to quantify progress. RAG Observability & Evals

Specialized metrics for retrieval-augmented generation pipelines including retrieval relevance, answer faithfulness, and context completeness.

Advanced Topics

Extend your evaluation approach for complex, multi-step applications.

Evaluating Multi-Turn Conversations

Evaluate coherence, memory, and resolution in conversational apps that maintain context across multiple turns. Simulated Multi-Turn Conversations

Generate and test user-AI exchanges safely before production. Build realistic dialogue scenarios for robust evaluation. Agent Evaluation

Assess end-to-end trajectories for agents with tool use and planning. Evaluate action selection and task completion. Vibe Coding a Custom Annotation UI

Build a custom annotation UI in minutes using Langfuse as data source Automatic Prompt Improvement with AI Agent Skills

Use the Langfuse skill to analyze trace feedback and iteratively improve your prompts. How to Calibrate your LLM-as-a-Judge with the Langfuse Skill

Test whether your judge prompt agrees with how you would label cases, and iterate until it does. Human-in-the-Loop Scoring

Capture human judgments from your own review tooling and ingest them into Langfuse as structured scores, validated against a score config.

Self-hosting

Export self-hosted Langfuse logs to ClickStack

Collect JSON stdout from langfuse-web and langfuse-worker and search it in ClickStack.

Cookbooks

Runnable cookbooks, sorted by topic. The notebook sources (ipynb) live in the cookbook directory of the repository.

This part of the page is interactive; see it on the original page.

Video Tutorials

This part of the page is interactive; see it on the original page.