Latitude
Open-source agent observability platform with prompt engineering and auto-evaluation.
Description
Short Intro: Latitude (latitude.so) is an open-source agent engineering platform founded in 2022 by César Miguelañez and Gerard Clos under the legal entity Latitude Data S.L. The platform began as a prompt management and evaluation tool before pivoting in 2025 toward agent-native observability — tracing multi-turn sessions, tool calls, and subagent activity in production. The platform is available as a managed cloud service and as a fully self-hosted deployment, with $1.32M in seed funding from CB Insights.
Key Capabilities:
Multi-turn agent tracing with full system prompt, tool call, and subagent visibility per session
PromptL, a proprietary prompt language with variables, conditionals, and loops for dynamic prompt logic
GEPA (Generative Evaluation from Production Annotations), which auto-generates evaluations from annotated production failures
Issue lifecycle tracking from active through in-progress, resolved, and regressed states
LLM-as-judge, programmatic, and human review evaluation with batch experiment support
AI gateway that deploys prompts as live API endpoints updated on each publish
AI coding agent skills for Claude Code, Cursor, and Codex to add tracing in one command
Cloud-hosted and self-hosted deployment options under a fully open-source license
Alternative tools
- OpenAI Playground
Browser-based prompt iteration environment for the OpenAI API.
- Galileo AI
Detect hallucinations and agent failures across the full development lifecycle
- LangWatch
Open-source LLMOps platform for observability, evaluation, and agent simulation.
- Adaline
End-to-end prompt management platform covering iteration, evaluation, deployment, and monitoring.
- Maxim AI
End-to-end AI evaluation platform with pre-production agent simulation and production observability
- Athina AI
Collaborative AI development platform for prototyping, evaluating, and monitoring LLM features.