Skip to content
@Shelflytics

Shelflytics

Shelflytics is a complete web to mobile solution that tracks, manages using AR technologies and embedded models with MCP to maximize product facings.

Shelflytics

Shelflytics is a complete web-to-mobile solution designed to track and manage product facings using cutting-edge Augmented Reality (AR) technologies, embedded ML models, and Model Context Protocol (MCP) to maximize product visibility and optimize in-store layouts.


📦 What is Shelflytics?

Shelflytics transforms physical retail environments through AR-powered mobile and web interfaces. It combines:

  • 📱 AR-based mobile scanning
  • 🧠 Embedded machine learning models
  • 🔁 Real-time sync via Model Context Protocol (MCP)
  • 📊 Analytics for product-facing performance

🧠 What is MCP?

Model Context Protocol (MCP), developed by Anthropic in late 2024, is an open-source, standardized JSON-RPC communication protocol. It enables AI models to seamlessly connect with external data sources and tools—such as your backend systems or product databases—without building custom integrations for each one :contentReference[oaicite:1]{index=1}.

MCP acts like a universal "USB-C port" for AI-enhanced apps, allowing your embedded models to access context (e.g., inventory databases, shelf-layout metadata) and perform actions via secure, interoperable connections :contentReference[oaicite:2]{index=2}.


🚀 Features

  • 🔍 AR Product Scanning
    Detect and analyze shelf layouts using your phone’s camera.

  • 🧠 Smart Layout Analytics
    On-device model assesses product-facing quality and placement.

  • 🔄 MCP Integration
    Embedded models retrieve real-time data and push updates using MCP.

  • 🌐 Web Dashboard
    View analytics, performance trends, and reports in real time.

  • 🔐 Secure Multi-user Management
    Roles and permissions for field agents, supervisors, and admins.


🖼️ Screenshots / Demo


📚 Documentation


🛠️ Tech Stack

Technology Purpose
React / Vue.js / Next.js Web Dashboard
Kotlin / Jetpack Compose Mobile AR App
TensorFlow Lite Embedded ML
ARKit / ARCore AR Tracking
FastAPI / Node.js Backend APIs
PostgreSQL / Supabase Data Storage + Authentication
Docker / Kubernetes Deployment & Scaling
MCP Context integration between models and backend systems

📋 Roadmap

  • Cloud-based model updates
  • Offline mobile scanning support
  • Shelf heatmap & insights
  • ERP system integration

Popular repositories Loading

  1. jake-sku-detector jake-sku-detector Public

    SKU Identifier trained on Retinanet Model

    Jupyter Notebook 2

  2. jake-doc-retriever jake-doc-retriever Public

    A REST API that manages SKU detection requests and provides additional features on top of the core RetinaNet detector.

    Python 2

  3. yolov11-mobile-model yolov11-mobile-model Public

    Repository that contains the code and model file for the YOLOv11 model to be embedded for live tagging.

    Jupyter Notebook

  4. mobile-app mobile-app Public

    Repository that contains the application code and deployable APK for the mobile app.

    Kotlin

  5. .github .github Public

  6. sophie-sales-predictor sophie-sales-predictor Public

    Jupyter Notebook

Repositories

Showing 8 of 8 repositories

People

This organization has no public members. You must be a member to see who’s a part of this organization.

Top languages

Loading…

Most used topics

Loading…