BuildWorld AI is a local-first visual systems-simulation lab for designing, running, comparing, and reporting graph-based models of complex systems. It focuses on bottlenecks, cascading failures, resilience, optimization, and emergent behavior while staying clear that the results are educational and exploratory—not certified predictions or professional advice.
Live demo · Case study · Architecture · Methodology · Testing
This repository is the authoritative source for the current standalone BuildWorld AI product. The AI Project Portfolio monorepo may contain a review copy, but product development, release evidence, and canonical documentation are maintained here on main.
Generated deployments, screenshots, and reports are outputs of this source and do not supersede the repository.
- Interactive Simulation Studio with draggable graph nodes, editable properties, run/pause/reset controls, speed control, and live metrics.
- Eight built-in demo scenarios: traffic, supply chain, power grid, ecosystem, warehouse, epidemic/population, emergency resilience, and a blank custom network.
- Deterministic simulation engines for flow networks, ecosystem dynamics, epidemic/population models, and cascade stress experiments.
- SSI System Stability Index: an original 0–100 score across throughput, bottleneck risk, resilience, redundancy, cascade resistance, resource balance, recovery, and optimization potential.
- Dashboard charts, event logs, critical-node analysis, optimization suggestions, snapshot comparison, Markdown/JSON report export, local save/load, project import/export, and print-to-PDF support.
- No paid AI API is required. The insight layer is deterministic and local.
- Reproducible reports record model version, seed, input fingerprint, and multi-seed uncertainty ranges.
- Named variants preserve baseline/alternative reasoning and expose changed graph inputs.
- React, TypeScript, and Vite
- Custom SVG/CSS visualizations
- Browser
localStoragepersistence - Vitest for pure simulation-logic tests
- Static deployment on Vercel, Netlify, or equivalent hosting
Prerequisite: Node.js 22 or later.
npm ci
npm run devOpen the local Vite URL shown in the terminal, usually http://127.0.0.1:5173.
npm run lint
npm run typecheck
npm run test
npm run buildThe GitHub Actions workflow runs the same checks on pushes and pull requests. Passing these checks establishes the recorded build and test result; it does not certify the model for engineering, public-health, infrastructure, financial, or safety-critical decisions.
- Small City Traffic Bottleneck
- Regional Supply Chain Disruption
- Neighborhood Power Grid Failure
- Forest Ecosystem Balance
- Warehouse Throughput Optimization
- Disease Spread in Connected Communities
- Emergency Shelter Resource Network
- Custom Blank Network
BuildWorld AI is not a certified engineering model, public-health tool, infrastructure-design recommendation system, ecological forecast, financial tool, or safety-critical decision system. It is intended for education, portfolio review, scenario reasoning, and product exploration. Its SSI, intervention suggestions, and multi-seed ranges are deterministic heuristics based on user-provided assumptions.
For a static deployment:
npm run buildDeploy the generated dist/ directory through Vercel for the canonical public release. The current live demo is deployed from main; record the source commit used for each production deployment so the public experience remains traceable. The static output can also be hosted elsewhere for private evaluation, but those hosts are not the current canonical deployment.