Products/Automation tools/Heym

Heym

Self-hosted AI workflow automation with agents, RAG, and MCP

Automation toolsBerlin, Germany2 peopleFounded 2026Visual canvas for workflow designMulti-agent orchestrationRAG pipelines (knowledge retrieval)HITL (Human-In-The-Loop) approvalsMCP server/client supportFull LLM traces with granular event trackingPortal to expose workflows as public chat UINo-code AI agent builderAutomatic context management for long-running agentsEvaluation system for prompt testing
Visit Heym →

Our Take

Okay, this one hits different. Heym is a self-hosted AI workflow builder that actually gives you agents, RAG, and MCP in one visual canvas—no glueing five tools together like some kind of workflow Frankenstein. The team is literally two people in Berlin running MIT+Commons Clause and they landed Day Rank #68 on Product Hunt with 66 upvotes, which is actually solid for a scrappy open-source launch. If you're an AI engineer tired of stitching random tools into a janky Rube Goldberg machine, this is the move. Free and self-hosted so you're not locked into another SaaS treadmill.

Build AI workflow automations visually, with no coding required. Heym supports multi-agent orchestration, RAG pipelines, and MCP, plus a powerful AI builder for faster creation.

Problem It Solves
Existing workflow platforms were built for simple, rule-based logic. When workflows became AI-native, users ended up gluing five tools together (agents, document retrieval, approval steps, observability) instead of building. Every time they needed agents, document retrieval, approval steps, and observability in the same flow, they ended up gluing five tools together.
Target Customer
Technical people building AI workflows, developers, AI engineers
Use Cases
Building AI workflow automations, Multi-agent orchestration, Document retrieval with RAG, Human review checkpoints in AI flows, Exposing workflows as AI assistant tools
Free Tier
true
Differentiator
Self-hosted, source-available (MIT + Commons Clause) with full LLM traces at every individual LLM call level, visual canvas, and no vendor lock-in
Why Now
As AI workflows become more complex with agents, RAG, and approval steps, existing tools force users to glue multiple platforms together. Heym offers all-in-one solution running on user's own infrastructure.
Traction
Notable Metrics: 66 upvotes, 25 followers, Day Rank #68

Key Facts

Category
Automation tools
Location
Berlin, Germany
Founded
2026
Team Size
2 people
Pricing
Free (source-available)
Discovered via
product-hunt

The people behind Heym

B

Burak Akgün

profile
C

Ceren Kaya Akgün

profile

Links

WebsiteGitHubSource: product-hunt

Want products like this in your inbox every morning?

Five products. Every morning. Written by someone who actually cares whether they're good or not. Free forever, unsubscribe whenever.