# Latitude

> An AI engineering platform for product teams to build, test, evaluate, and deploy reliable AI agents and prompts.

Latitude is an AI engineering platform designed for product teams to build, test, evaluate, and deploy reliable AI agents and prompts. The platform provides a comprehensive suite of tools for prompt management, experimentation, observability, and deployment, enabling teams to ship AI products with confidence and full visibility into how their agents think and perform.

- **Prompt Manager** allows teams to design and test prompts at scale before shipping to production, with advanced templates for complex scenarios and the ability to connect integrations or custom tools.

- **Evaluation System** supports LLM-as-judge, human-in-the-loop, and ground truth evaluations using production or synthetic data, enabling teams to run experiments in batch and compare results.

- **Version Control** keeps track of every change with the ability to publish new prompts directly from Latitude and push updates to production smoothly.

- **Observability** provides full monitoring and debugging capabilities for agents and prompts in production, with real-time performance tracking, complete logs, and version comparison.

- **Latte AI Assistant** helps build agents by understanding context and providing cross-project suggestions based on what worked before.

- **Extensive Integrations** with over 2,800 two-way integrations, supporting custom tools, APIs, internal systems, and MCPs to let agents read data and take action.

- **Collaboration Workspaces** enable teams to share agents, experiments, and results in one place, keeping everyone aligned on agent performance.

- **SDK and API Integration** allows seamless integration into production environments using TypeScript SDK or direct API access.

To get started, sign up for a free account on the Latitude platform, design your prompts using the prompt manager, run evaluations to test performance, and deploy with confidence using version control. The platform supports both cloud deployment and self-hosting options for maximum flexibility and privacy.

## Features
- Prompt manager with advanced templates
- LLM-as-judge evaluations
- Human-in-the-loop evaluations
- Ground truth evaluations
- Version control for prompts
- Real-time observability and monitoring
- Complete logging and debugging
- Batch experiments
- AI assistant (Latte) for building agents
- 2,800+ two-way integrations
- Custom tools and API support
- MCP support
- Team collaboration workspaces
- TypeScript SDK
- API integration
- Self-hosting option
- Cloud deployment

## Integrations
Custom APIs, MCP servers, TypeScript SDK

## Platforms
WEB, API, DEVELOPER_SDK

## Pricing
Open Source, Free tier available

## Links
- Website: https://latitude.so
- Documentation: https://docs.latitude.so/
- Repository: https://github.com/latitude-dev/latitude-llm
- EveryDev.ai: https://www.everydev.ai/tools/latitude