Welcome to Artanis
Artanis is an observability platform for AI applications that helps you understand failures, manage evaluation sets, and act on user feedback—all through comprehensive tracing.Core Capabilities
- Error Analysis - For each labeled trace, we automatically determine the reason for discrepancy, categorizing errors to help you understand patterns and prioritize fixes.
- Searchable Traces - Every trace is captured and fully explorable. Search by inputs, outputs, metadata, or errors to quickly find what you need.
- State Management - We capture all state at inference time, so you can update your model and replay the exact same data to verify improvements.
- Trace → Evals - Easily label any trace and add it to evaluation sets. Export in formats compatible with your existing evaluation scripts.
- Slack Feedback Mapping - When customers send feedback via Slack, our bot automatically matches messages to existing traces—no manual linking required.
Quick Start
Get tracing working in under 5 minutes.Installation
Install the SDK for Python or TypeScript
Quick Reference
One-page cheat sheet for common operations
API Reference
Complete API documentation
Tracing Guide
Learn the tracing pattern in depth
Common Use Cases
See how teams use Artanis to improve their AI applications.RAG Pipeline Analysis
Trace retrieval quality and generation accuracy
Classification Debugging
Identify misclassification patterns
Feedback Integration
Connect feedback to traces automatically
Error Handling
Understand how the SDK handles errors