01 / CLIENTS
Applications
Electron · React
Flutter · QR scan
Back to experience
April 2025 · Construction Industrielle Africaine (C.I.A) · Casablanca, Morocco
Internship Project: C.I.A STOCK
Complete real-time and offline inventory management ecosystem
Designed and developed C.I.A STOCK, an inventory management system for industrial consumables, bringing desktop administration and mobile stock operations into one solution.
Architecture
System architecture
The project is based on an asynchronous distributed client-server architecture, built for maximum resilience against internet instability typical in remote workshops and work sites.
02 / NETWORK
Secure network
Connects the mobile app to the local PC server over LAN or VPN.
TailscaleWireGuard
03 / API
Application server
The API receives requests and applies inventory business rules.
PythonFastAPI
04 / DATA
Database
Stores products, stock movements, users, and configuration.
MongoDBMotor
05 / RESPONSE
Results return to the interfaces
After persistence, the desktop app refreshes stock; the report describes synchronization to the interface via SSE.
DATA PIPELINE
From scan to stock update
When a device is connected, each stock movement follows these steps.
STEP 01
Capture
The mobile app scans the product QR code and prepares the stock operation.
STEP 02
Transport
The request reaches the API over the local network or Tailscale VPN.
STEP 03
Processing
FastAPI validates the operation, checks business rules, and calculates product status.
STEP 04
Persistence
MongoDB records the product or stock movement through the asynchronous Motor driver.
STEP 05
Update
The desktop interface reflects the new stock state; synchronization is described via SSE.
OFFLINE PATH
When the network is unavailable
The report describes local storage on mobile: operations are kept offline and synchronized when the connection returns.
Overview
Internship Project: C.I.A STOCK
The C.I.A STOCK project consists of the design and development of a complete software ecosystem for inventory management at Construction Industrielle Africaine (C.I.A). The primary goal is to modernize technical consumables management (industrial gas, welding electrodes, tools) which was previously manual, in order to eliminate stock-outs that paralyze heavy industrial work sites (mines, cement plants, civil engineering sites).
The system shifts the company from a traditional paper-based flow to a smooth, secure, and zero-paper digital solution.
PERSISTENCE
Managed data
The report identifies four main MongoDB collections:
- Products & thresholds
- Stock movements
- Users & roles
- Configuration
Tech Stack
Tech Stack Breakdown
01
Desktop application
Cross-platform desktop console built with Electron, React (Vite), and Tailwind CSS for administration, inventory dashboards, and product management.
ElectronReactViteTailwind CSS
02
Mobile application
Cross-platform mobile application development (Android/iOS) with a single codebase for native performance on the field.
Flutter
03
Backend API
High-performance async backend utilizing Python async/await to handle load peaks and auto-generate OpenAPI/Swagger documentation.
PythonFastAPI
04
Database
NoSQL database selected for schema flexibility, ideal for indexing highly diverse technical items with different specifications.
MongoDBMotor
05
Secure connectivity
Tailscale VPN connects the mobile terminal to the local PC server; communications are encrypted with WireGuard.
TailscaleWireGuard
06
Document generation
Automated generation of branded C.I.A purchase orders in PDF format using Python.
PDFPython
Innovations & AI
Innovations & AI
01
Resilience & Offline Mode
Local storage cache enabling inventory scanning and registration offline, with automatic synchronization upon reconnecting.
02
Industrial Ergonomics
Dark mode designed for dim warehouses and large incremental touch buttons usable with protective work gloves.
03
Zero Trust Mesh Security
End-to-end encrypted tunnels. Simple QR code scan configuration for field operators to pair devices instantly.
04
Smart Critical Thresholds
Real-time stock status calculations (Optimal, Critical, Out of Stock) with automatic check-out blocking and urgent alert systems.
05
AI & Industry 4.0 Vision
Conceptualized AI integrations: Time-series demand forecasting (LSTM/Prophet), computer vision visual item identification (YOLO/ResNet), predictive maintenance on heavy machinery, and LLM-powered natural language inventory assistant.