Skip to main content

Michelangelo AI UI

Michelangelo AI's User Interface (UI) provides a standard, code free ML development experience. It guides users through five phases in the ML development lifecycle, from developing your first model to productionization.

Getting Started​

Deploy to Kubernetes​

Production deployment using the Michelangelo AI Helm chart (and its sandbox values) as a template for your infrastructure.

  • For: Platform operators
  • Use case: Full ML platform deployment with UI

→ Deploying Michelangelo AI UI

Integrate with Existing React App​

Add Michelangelo AI components to your existing React application as npm dependencies that connect to your infrastructure.

  • For: Frontend developers, application teams
  • Use case: Separate frontend/backend infrastructure, or embedding ML capabilities in existing developer tools

→ React Library Integration

Local Development​

Set up a development environment for contributing to the UI codebase.

  • For: Contributors, UI developers
  • Use case: UI development and contributions

→ Local Development Setup

Architecture​

The Michelangelo AI UI is built with React and communicates with the Michelangelo AI API server through an Envoy proxy that transcodes proto3 JSON requests to gRPC. The UI supports two main consumption methods:

  • Containerized deployment: Complete UI deployed to Kubernetes clusters
  • Component integration: Individual React components embedded in existing applications

For internal architecture documentation (providers, hooks, types, and component patterns), see the UI Developer Reference.