Latitude — AI Agent Engineering Platform

Introduction

Latitude is an open-source AI agent engineering platform with 4,000+ GitHub stars. It provides full observability for LLM applications — capturing prompts, inputs/outputs, tool calls, and performance metrics. Supports a Prompt Playground, dataset curation, LLM-as-judge evaluation, cross-model experiments, and automated evaluation guards. Ideal for teams building production-grade AI apps.

Latitude — Build, Evaluate, and Monitor AI Agents

Core Features

• Observability — capture prompts, I/O, tool calls, latency, cost
• Prompt Playground — iterate prompts with instant feedback
• Datasets — curate test data
• Evaluation — LLM-as-judge and custom metrics
• Experiments — compare performance across models
• Auto-guards — automatic checks before responses go out

FAQ

Q: What is Latitude? A: An open-source AI agent engineering platform providing observability, evaluation, and monitoring.

Q: How does it differ from LangFuse? A: Latitude covers the full agent engineering lifecycle with built-in LLM-as-judge and experiment comparisons.