Heym

What Heym is

Heym is infrastructure for building, running, and governing AI agents and workflows in production. AI, platform, and automation teams can create workflows on a visual canvas or describe them in natural language or voice for the assistant to generate. An orchestrator can delegate tasks to named sub-agents and sub-workflows, with nesting up to five levels. Retrieval uses managed Qdrant or built-in Postgres with pgvector, supporting document uploads and semantic search; Cohere reranking is optional. Human-review checkpoints can pause a workflow while a reviewer accepts, edits, or refuses the work before it resumes. Execution traces record prompts, responses, tool calls, timing, and token use, and the software includes evaluation features. Listed integrations include OpenAI, Anthropic, Google Gemini, Slack, Discord, Gmail, Outlook, GitHub, Jira, Telegram, Notion, Qdrant, Redis, RabbitMQ, Google Sheets, and MCP. Teams can deploy Heym on infrastructure they control using Docker Compose or Kubernetes. The features page describes credentials encrypted at rest with AES-256 Fernet and JWT authentication using HttpOnly cookies and refresh token rotation. Heym is source-available under Commons Clause plus MIT. The site says it can be used, modified, and distributed without licensing fees, but it cannot be resold as a paid service. Community help is available through Discord; support packages and enterprise engagements may be paid services.

Who it suits

Heym is aimed at AI, platform, and automation teams building and operating agent workflows. It suits teams that want self-hosted deployment, workflow orchestration, retrieval, human review, and execution traces on infrastructure they control.

Limits

The Free to Use plan has no licensing fees or usage limits and is self-hosted. The software cannot be resold as a paid service under its Commons Clause plus MIT license.

Getting started

1. Open the Heym website.
2. Choose Docker Compose or Kubernetes for deployment on infrastructure you control.
3. Build a workflow on the visual canvas or describe one in natural language or voice.
4. Connect the services and retrieval options needed from Heym’s listed integrations and supported stores.
5. Add human-review checkpoints or delegate work to named sub-agents and sub-workflows as needed.

Verdict

Choose Heym if your AI, platform, or automation team wants to build and govern workflows on infrastructure you control, with human review and detailed execution traces. Look elsewhere if you need to resell the software as a paid service; its Commons Clause plus MIT license disallows that.