Skip to main content

Assistants Evaluation (Langfuse)

This comprehensive guide explains how to install and configure Langfuse using Helm, with both automated and manual deployment methods.

Overview​

Langfuse is an open-source LLM observability platform that provides:

  • Tracing: Track and analyze LLM calls and their performance
  • Evaluation: Assess and score AI assistant responses
  • Analytics: Gain insights into usage patterns and costs
  • Debugging: Identify and troubleshoot issues in LLM applications

Deployment Options​

This guide provides two deployment methods:

Uses the deploy-langfuse.sh script to automatically handle:

  • Kubernetes secret creation
  • Helm repository configuration
  • Langfuse deployment
  • Integration secret creation for CodeMie

See Deployment for both automated and manual deployment options.

Documentation Structure​

Follow these sections in order for a successful deployment:

  1. Prerequisites - Required tools and infrastructure
  2. System Requirements - Resource specifications and architecture
  3. Deployment Prerequisites - Configuration steps before deployment
  4. Deployment - Automated or manual deployment options
  5. Post-Deployment Configuration - Configure CodeMie integration
  6. Troubleshooting - Common issues and solutions

Next Steps​

Start with Prerequisites to ensure your environment is ready for Langfuse deployment.