Hi, I'm Matt York — engineer, blockchain researcher, and creator of pflow.xyz, a Petri net toolkit for modeling complex systems.
pflow — an ecosystem of tools that treat Petri nets as a universal abstraction for state machines, workflows, and verifiable computation.
| Project | What it does |
|---|---|
| pflow-xyz | Browser-based visual editor and ODE simulator |
| go-pflow | Go library — modeling, simulation, process mining, ZK proofs |
| petri-pilot | MCP server for AI-assisted design and deterministic full-stack codegen |
| pflow-jl | Julia framework for Petri net visualization and analysis |
Try it: 18 interactive demos — games, workflows, scientific models, and more.
A Petri net model defines places (states), transitions (actions), and arcs (connections). From that single JSON-LD file:
- go-pflow simulates it — discrete-event, continuous-time ODE, or ZK proof generation
- petri-pilot generates a full-stack app — Go backend, ES modules frontend, GraphQL API
- pflow-xyz lets you build and edit it visually in the browser
The LLM designs models. Templates produce apps. No LLM-generated code in the output.
Petri nets provide transparent, deterministic logic — the model itself is the explanation. I'm working to make them practical and accessible through open tools, minimal dependencies, and visual interfaces.
- Vanilla JS/ES modules, Go backends, SQLite — no framework bloat
- Dual execution (Go + JS) for verifiable correctness
- Content-addressed storage (IPFS CIDv1) for immutable models
- ZK proofs for trustless state transition verification
- Visual Editor: pflow.xyz
- Interactive Demos: pilot.pflow.xyz
- Blog: blog.stackdump.com
- Organization: github.com/pflow-xyz