Langfuse

Langfuse is an open-source LLM engineering platform that helps teams build, monitor, and improve AI applications. It covers the full development lifecycle with tracing, prompt management, evaluations, and analytics dashboards in one place.

Key Features

• LLM Observability: Trace and debug generative design features in production.
• Prompt Management: Manage prompts through natural language via CLI, MCP, or coding agents.
• Evaluation: Set up evaluations to measure and improve application performance.
• Metrics & Analytics: Dashboards for monitoring and analyzing LLM application behavior.
• Agent Integration: Works with Claude Code, Cursor, Codex and other coding agents via CLI and MCP servers.
• Self-Hosting: Available for self-hosted deployment with ClickHouse backend.

Use Cases

• Debugging production LLM applications
• Managing prompts across development lifecycle
• Evaluating LLM outputs and agent performance
• Monitoring and improving AI-powered features

Who It's For

• AI teams building production-grade agents and LLM applications
• Developers who value open-source and control over their data
• Organizations requiring enterprise security and compliance

Workflow

Langfuse powers the entire development cycle from prototype to full scale production loads. Tracing never blocks your application with async ingestion and automatic batching.

Key Benefits

• Open-source with MIT license
• Full development lifecycle coverage (tracing, prompts, evals, metrics)
• Integrates with coding agents (Claude Code, Cursor, Codex)
• Enterprise scale with ClickHouse backend and SOC 2 / ISO 27001 compliance