Cookbook
Practical AI observability recipes - tracing patterns, integration guides, evaluation strategies, and production monitoring examples
Cookbook
Learn AI observability through step-by-step tutorials and copy-paste recipes. Whether you're just getting started or looking to solve specific problems, you'll find practical guidance here.
Tutorials
Complete, step-by-step guides for implementing AI observability in real-world scenarios.
RAG Application Tracing
Add full observability to retrieval-augmented generation pipelines
Agent Evaluation
Evaluate AI agents with tool use and multi-step reasoning
Cost Optimization
Reduce AI costs while maintaining quality
Production Monitoring
Set up monitoring, alerts, and dashboards for production AI
Tracing Recipes
Copy-paste solutions for common tracing patterns.
Async Tracing
Trace asynchronous and concurrent AI operations
Batch Processing
Trace batch LLM operations efficiently
Streaming Responses
Trace streaming LLM outputs in real-time
Evaluation Recipes
Patterns for measuring and improving AI quality.
Hallucination Detection
Detect and measure hallucinations in LLM outputs
Response Quality
Measure response quality across multiple dimensions
Integration Recipes
Framework-specific implementation patterns.
Next.js Application
Add tracing to a Next.js AI application
FastAPI Service
Add tracing to a FastAPI backend service
Content Format
Tutorials follow a consistent structure:
- Prerequisites - What you need before starting
- Overview - What you'll build and learn
- Step-by-step implementation
- Testing & validation
- Next steps
Recipes are self-contained solutions:
- Problem - What you're trying to solve
- Solution - Complete, working code
- How it works - Key concepts explained
- Variations - Alternative approaches
Contributing
Have a tutorial or recipe to share? Open a pull request or start a discussion.