heymrun/heym

What it solves

Heym is an AI-native automation platform that allows users to build, visualize, and run intelligent workflows without writing code. It addresses the complexity of orchestrating multiple AI agents, RAG pipelines, and external tools, providing a visual canvas for design and deployment.

How it works

Users create workflows on a drag-and-drop canvas using a library of nodes, including LLM and Agent nodes. The platform supports multi-agent orchestration where a primary agent can manage sub-agents and sub-workflows in parallel. It integrates with vector stores (Qdrant or Postgres/pgvector) for RAG, supports the Model Context Protocol (MCP) for tool connectivity, and allows for human-in-the-loop checkpoints to pause and approve agent actions.

Who it’s for

It is designed for developers and enterprise teams who want to build and deploy production-ready AI automations that are observable, scalable, and self-hostable.

Highlights

• Visual Workflow Editor: A drag-and-drop canvas for building AI-native automations.
• AI Assistant: Generates and wires nodes on the canvas automatically from natural language or voice prompts.
• Multi-Agent Orchestration: Supports parallel execution of sub-agents to reduce latency and context bloat.
• Built-in RAG: Native support for managed vector stores for semantic search.
• MCP Support: Connects agents to MCP servers or exposes workflows as MCP servers for other clients.
• Observability: Full LLM traces, real-time USD cost tracking, and OpenTelemetry export.
• Agentic Kanban Board: Persistent agentic jobs where moving cards triggers specific workflow chains.
• Human-in-the-Loop: Ability to pause agent execution for user approval or input.