-blue?style=flat-square){kind=icon}
Official public release distribution, technical documentation, and architecture showcase for the Vectoris Engineering Workstation.
Vectoris is an advanced engineering workstation and project intelligence platform engineered for electrical estimators, MEP (Mechanical, Electrical, Plumbing) coordinators, and preconstruction teams. Combining a local-first desktop runtime with vision-language neural perception models, Vectoris automates the labor-intensive discipline of construction drawing takeoff, quantity calculation, and Bill of Quantities (BOQ) synthesis directly from native architectural drawings without exposing confidential project IP to uncontrolled third-party cloud infrastructure.
Vectoris unifies preconstruction workflows into a structured Project Intelligence Hierarchy, transitioning from native blueprints to procurement-ready bids:
VECTORIS WORKSPACE
│
┌─────────── PROJECT ───────────┐
│ │
PROJECT INTELLIGENCE │
│ │
┌───────────┼────────────┐ │
↓ ↓ ↓ ↓
Drawings Documents AI Models Collaboration
│
↓
Takeoff (Linear Runs & Symbol Counts)
│
↓
Bill of Quantities (BOQ Synthesis)
│
↓
Engineering (Conductor & Conduit Sizing)
│
↓
Estimation (Labor Units & Material Pricing)
│
↓
Commercial Bidding & Proposal Delivery
- Local-First Blueprint Takeoff: On-device vector extraction and layer segregation for multi-megabyte CAD DWG, DXF, and PDF sheets without cloud latency or security risk.
- Multimodal Neural Symbol Perception: High-accuracy classification of electrical receptacles, switchboards, luminaires, and low-voltage drops with a 99.1% benchmark precision rate.
- Human-in-the-Loop Verification Queue: Staged inspection queue with bounding overlays, confidence tiers, and batch approval tools ensuring zero "black box" estimation errors.
- Automated NEC Feeder & Conduit Sizing: Live AI engineering copilot calculating voltage drop across long feeder runs (e.g., proposing 350 kcmil conductor sizing for 184.6-meter runs) and raceway fill capacity.
- Tauri v2 Native Desktop Shell: Rust 2021 native core with hardware-accelerated WebView2, custom frameless window management, and sub-95MB idle RAM footprint.
- Liquid Theme System: 5-phase fluid Bézier wave compositor transitions between deep dark mode (black cherry/coffee bean) and alabaster light mode.
- Cryptographic Trust Chain: Minisign Ed25519 signature verification on all release binaries, featuring the atomic "Stay Put" update handoff.
| Metric | Benchmark Result | Industry Comparison |
|---|---|---|
| Sheet Ingestion & Vectorization | < 1.2 seconds (50MB drawing package) | 8–15s (Cloud-based SaaS) |
| Idle Memory Footprint | < 95 MB RAM | 850MB–1.4GB (Electron runtimes) |
| Takeoff Detection Throughput | 18,000+ items across 400+ sheets | Days of manual tracing |
| Perception Precision | 99.1% across standard MEP schedules | Variable human error |
| Offline Reliability | 100% local compute availability | Fails without active internet |
Signed production installers and deployment packages are published on GitHub Releases.
| Package | Target Platform | Description |
|---|---|---|
Vectoris_0.2.5_x64-setup.exe |
Windows 10/11 (x86_64) | Standard per-user NSIS interactive installer |
Vectoris_0.2.5_x64_en-US.msi |
Windows 10/11 (x86_64) | Enterprise MSI package for silent IT / GPO / Intune deployments |
latest.json |
All Windows Workstations | Cryptographically signed update manifest for in-app updater |
Every Vectoris software release is cryptographically signed using Ed25519 asymmetric keys in Minisign format.
dW50cnVzdGVkIGNvbW1lbnQ6IG1pbmlzaWduIHB1YmxpYyBrZXk6IDVBM0NBNEIzQzFERDgxRjYKUldUMmdkM0JzNlE4V2xiaXJaZWkyWThFUFVYSDhMV0JROGl3T053V2FJNnFKVWl0L3hkNW4zNVEK
To verify a downloaded installer before execution:
minisign -Vm Vectoris_0.2.5_x64-setup.exe -p vectoris.pubFor complete verification steps and SHA-256 validation, see VERIFICATION.md.
Explore the comprehensive documentation suite for deep architectural and operational details:
| Document | Description |
|---|---|
| 🏛️ System Architecture | Deep technical blueprint: Tauri v2, Rust native FFI, React 19, and IPC service boundaries |
| 📐 Project Intelligence | Core domain model: Project containers, linear takeoffs, BOQ synthesis, and bidding |
| 🧠 AI Perception Pipeline | Neural vision symbol classification, vector geometry extraction, and NEC sizing |
| 🎨 Liquid Design System | 5-phase fluid Bézier wave compositor, dual color palettes, and workstation ergonomics |
| 💻 Installation & Deployment | System requirements, NSIS/MSI deployment instructions, and troubleshooting |
| ❓ Frequently Asked Questions | Answers on local-first privacy, drawing formats, offline operation, and accuracy |
| 🗺️ Product Roadmap | Strategic horizons: Revision diffing, spec parsing, and enterprise ERP connectors |
| 🛡️ Security Policy | Vulnerability reporting protocol, supported versions, and runtime isolation |
| 🔐 Release Verification | Guide for manual and automated cryptographic Ed25519 signature checks |
| 🤝 Contributing Guidelines | Architectural guardrails, local development toolchain, and quality standards |
| 📜 Code of Conduct | Engineering standards of technical craft, mathematical integrity, and ethics |
| 📄 License | Proprietary and confidential enterprise software license |
Vectoris is developed and maintained by Vectoris AI Inc.
- Public Distribution Repository:
VectorisAI/Vectoris - Primary Upstream Repository:
HardikBhaskar2010/Vectoris - Security Inquiries:
security@vectoris.ai - Engineering Contact:
engineering@vectoris.ai
Proprietary Enterprise Software. Copyright © 2026 Vectoris AI Inc. All rights reserved.
Unauthorized copying, reverse engineering, decompilation, or redistribution is strictly prohibited. See LICENSE for terms.