Engineered for Apple Silicon (M4/M1) | Powered by MLX & MPS
JuniorOmega is a high-fidelity, low-power spatial data orchestration engine. It bridges the gap between raw edge-sensor ingestion (TrueDepth/LiDAR), manifold SVD topology generation, and automated fabrication (G-code). By leveraging Apple’s mlx.core, JuniorOmega executes high-density tensor math on-device, eliminating cloud latency and traditional PyTorch overhead.
The SDK is organized into five sovereign nodes, now expanded for decentralized edge compute:
- 1. Ingestion (Omni-Vision): Multi-modal data capture across iPad Pro (TrueDepth), MacBook (FaceTime HD), and OAK-D spatial microprocessors.
- 2. Transport (gRPC/NIO): A decentralized "Nervous System" using gRPC to stream 1220-vertex facial/spatial manifolds from Swift-based mobile nodes to the M4 compute core.
-
3. Core (MLX SVD): Executes native Anisotropic Tensor Decomposition (
$A = U \Sigma V^T$ ) to extract surface normals, curvature, and topological stability. - 4. Sandbox (Mid-Engine): An interactive layer for BitNet/LLM agentic overrides, allowing real-time adjustment of anomaly thresholds before fabrication.
- 5. Fabrication (G-Code): Translates optimized meshes into topology-aware, 3-axis G-code toolpaths using adaptive feedrate and spline interpolation.
JuniorOmega now includes a TradingView-style Web Dashboard for real-time monitoring:
- Live SVD Audit: View stability metrics and curvature coefficients.
- 3D Manifold Preview: Interactive Plotly-driven 3D visualization of incoming point clouds.
- Metal Rendering: Headless Blender integration for high-contrast "Sovereign Gold" artifact shading.
- Hardware: Apple Silicon (M4/M3/M2/M1) | iPad Pro/iPhone (for TrueDepth features).
- OS: macOS 14.0+ (Sonoma/Sequoia) | iOS 17.0+.
- Python: 3.10+ (Optimized for 3.12).
- Key Dependencies:
mlx,grpcio,gradio,scipy,trimesh,mediapipe.
git clone https://github.com/JuniorCloudLLC/JuniorOmega.git
cd JuniorOmega
python3 -m venv .venv
source .venv/bin/activate
python3 -m pip install -r requirements.txtTo start the gRPC receiver (WOS10 Edge Node) and the Command Center:
# Terminal A: Start the gRPC Listener
python3 omega_orchestrator.py --grpc
# Terminal B: Start the Web Dashboard
python3 ui_controller.pyNavigate to swift_app/ and open Package.swift in Xcode. Deploy to a TrueDepth-capable iPhone/iPad to begin streaming spatial data to your local node.
JuniorOmega utilizes the Enterprise Harvester for public releases. This ensures:
- IP Protection: Proprietary MLX math kernels are automatically scrubbed and replaced with Community Stubs in public branches.
- Vaulting: Non-manifest artifacts and test data are moved to
Vaults/Quarantineto maintain repository hygiene. - Auditability: Every harvest generates a JSON audit log for regulatory and engineering compliance.
Identity: Generated for the Lead Architect of JuniorCloud LLC.
Tone: Logic-dense engineering rigor. No fluff. Finalized for Public Release.