# Maxim

> Enterprise-grade AI evaluation and observability platform for testing, monitoring, and improving AI agents and LLM applications.

Maxim is an enterprise-grade AI evaluation and observability platform that empowers developers to ship AI applications with quality, reliability, and speed. It provides end-to-end tooling for prompt experimentation, agent simulation, online evaluation, and production observability. Built by a team with backgrounds at Google, Slack, and Postman, Maxim serves as the missing quality layer for modern AI applications.

- **Prompt Playground** — *Experiment with prompts, compare outputs side-by-side, and version/deploy prompts directly from the UI.*
- **No-Code Agent Builder** — *Build and test AI agents without writing code using a visual interface.*
- **Agent Simulation & Evaluation** — *Run single and comparison agent simulations, evaluate voice agents, and schedule automated runs to catch regressions.*
- **Evaluator Store** — *Access Maxim's built-in evaluators or create custom ones; supports human evaluation workflows and managed human evaluation on Enterprise.*
- **Production Observability** — *Capture logs and traces from production, apply advanced filtering, and run online evaluations on live data.*
- **Dataset Management** — *Create datasets from production logs, manage entries, and use them to drive evaluation pipelines.*
- **CI/CD Integrations** — *Plug evaluations into existing CI/CD pipelines to enforce quality gates before deployment.*
- **Custom Dashboards & Reports** — *Build live dashboards and comparison reports to track model and agent performance over time.*
- **PII Management & RBAC** — *Protect sensitive data in logs and control access with role-based permissions.*
- **Broad Framework Integrations** — *Connect via SDK to LangChain, LangGraph, OpenAI, CrewAI, LiteLLM, Anthropic, Bedrock, Mistral, LiveKit, and more.*
- **Enterprise Security** — *SOC 2 Type II, ISO 27001, HIPAA, GDPR compliance; supports SAML SSO, In-VPC deployments, audit logs, and custom BAAs.*

## Features
- Prompt playground
- Prompt versioning and deployment
- No-code agent builder
- Agent simulation and evaluation
- Voice agent evaluation
- Scheduled evaluation runs
- Online evaluation on production data
- Custom evaluators
- Human evaluation support
- CI/CD integrations
- Production logs and traces
- Advanced log filtering
- Dataset creation from logs
- PII management
- RBAC with custom roles
- Custom dashboards
- Comparison reports
- SAML SSO
- In-VPC deployments
- SOC 2 Type II compliance
- ISO 27001 compliance
- HIPAA compliance
- GDPR compliance

## Integrations
LangChain, LangGraph, OpenAI, OpenAI Agents SDK, LiveKit, CrewAI, Agno, LiteLLM, LiteLLM Proxy, Anthropic, AWS Bedrock, Mistral

## Platforms
WEB, API

## Pricing
Open Source, Free tier available

## Links
- Website: https://www.getmaxim.ai
- Documentation: https://www.getmaxim.ai/docs
- Repository: https://github.com/maximhq
- EveryDev.ai: https://www.everydev.ai/tools/maxim-ai