langfuse.com

Use the Langfuse data model correctly: traces vs observations vs spans vs generations vs events, client-generated IDs, and async ingestion

Self-host Langfuse v3 with Docker Compose: required components, secrets, first-run steps, and production warnings

Track LLM token usage and cost accurately in Langfuse, including custom model prices for unknown models and manually supplied usage

Send traces from any OpenTelemetry-instrumented application to Langfuse's OTLP endpoint (exact path, auth headers, env vars)

Authenticate to and query the Langfuse Public API across EU/US/JP/HIPAA regions, with pagination and rate-limit handling

Configure an LLM-as-a-judge evaluator in Langfuse to automatically score production traces (model, variable mapping, filters, sampling)

Create a Langfuse dataset and dataset items, then run a dataset experiment and compare runs across prompt or model versions

Attach user feedback and custom quality scores (numeric, categorical, boolean, text) to a Langfuse trace or observation via the Scores API and SDK

Link a Langfuse generation to the exact prompt version that produced it, enabling cost and quality comparison across prompt versions

Fetch and compile a versioned Langfuse prompt at runtime using labels, client-side caching, and a fallback for outage resilience

Redact PII from Langfuse traces with a masking function and control trace sampling before data leaves the application

Group Langfuse traces into sessions, attribute them to end users, and add tags and metadata for filtering in the UI

Instrument a TypeScript/JavaScript app with the Langfuse JS/TS SDK v5 (OpenTelemetry-based), including flushing in serverless and edge runtimes

Wrap OpenAI LLM calls with the Langfuse Python drop-in integration to auto-capture tokens, cost, model params, and latency

Instrument a Python app end-to-end with the Langfuse Python SDK v4 (install, env vars, client init, auth check, @observe, nesting, flush)