just shipped ynode - an open-source visual workflow automation platform. Think n8n, but lighter and more hackable.
What it does:
- Visual node-based workflow builder (React Flow + TypeScript)
- Create custom nodes via CLI (and using Node Assembler - in app editor)
- Full-stack: vite as frontend, Express backend, bettersqlite storage
- Extensible architecture with pnpm monorepo
Tech stack:
- Frontend: React 19 + Vite + React Flow
- Backend: Node.js + Express + WebSocket
- Database: SQLite with credential encryption
- TypeScript throughout
Why I'm sharing:
I'm curious if this approach - building products through AI collaboration rather than traditional coding - has real market value. Can you actually:
- Get hired as a "vibe-coder"?
- Build profitable products this way?
What I'm looking for:
Honest feedback on the codebase - is it maintainable? Would you use it?
Use cases - what would you actually automate with this?
Direction - should I focus on making this a product, or is it just a portfolio piece?
The repo is here: https://github.com/iamyureka/ynode
Try it, break it, fork it. I want to know if this is actually useful or just another side project collecting dust.
Current state:
- Fully functional local setup
- Custom node generation via CLI
- Clean monorepo architecture
Questions for the community:
- Have you monetized projects built primarily with AI assistance?
- What makes a workflow automation tool worth paying for?
- Am I solving a problem that actually exists?
- Are only can/have vibe coding skill actually worth to get a job?
Real talk: I can ship products fast this way, but I'm not sure if that translates to actual opportunities. Feedback welcome.