ElderLink is a health and safety platform for elderly users and caregivers. Flutter mobile app with panic button, medicine reminders, health monitoring, and optional smartwatch-style UI. Node.js/Express backend with MongoDB; admin dashboard for staff, roles, and logs. Built with Flutter, Dart, Node.js, Express, MongoDB, and REST APIs.

Current capabilities (high level)
Backend

Elders, medicines, readings, heart alerts, music sessions
Strict MongoDB ObjectId handling and elder resolution for API calls
Medicine-related persistence and outbox-style event helpers (see backend/models/medicineEvent.js, backend/services/medicineEventOutbox.js)
Mobile (mobile/)

Firebase Authentication for staff (email/password and related flows)
Admin / staff areas: live data, logs, roles, settings
Elders and medicines CRUD against the Node API
Backend connection: host/port stored in app settings (defaults + --dart-define=MOBILE_API_HOST / MOBILE_API_PORT); see mobile/lib/screens/backend_settings_screen.dart
Alerts, music, profile / privacy screens
Watch (watch/)

Radial home: Medicine, Switch (change active resident), Clock, My Info, Music, Settings, Health; panic in the center
Medicine schedule from API; Karachi wall-clock for “today” scheduling
Switch resident: recent residents plus facility list from GET /api/elders merged into device history
Cross-elder banner: if another resident has pending doses today, optional prompt to switch
Heart rate / BP style monitoring UI, readings API, panic flow, music with session reporting (when enabled)

Desktop application for real-time weight monitoring and fabric roll ticket generation. Built with React, Electron, and Vite. Connects to weighing indicators via serial port, shows live weight, generates QR-coded tickets, tracks weight history, and supports printing + backups.

Highlights

Realtime weight monitoring + stable value detection
Serial port auto-scan + error handling
Ticket generation with QR codes + print flow
Languages: JavaScript

Features
Desktop Application
Real-time weight monitoring from YH-T7E Weighing Indicator
Auto-scan serial ports with multiple baud rate testing
Stable value detection (5 consecutive readings)
Fabric roll ticket creation with professional layout
QR code generation with all ticket data
Weight history tracking and management
Google Drive integration for data backup
Print functionality with formatted output
Python Serial Connection
Automatic Port Scanning: Scans all available serial ports to find the connected device
Real-time Weight Display: Continuously displays weight readings from the device
Status Monitoring: Detects when the machine is ON, OFF, or disconnected
Configurable Settings: Adjustable baudrate and timeout settings
Robust Error Handling: Graceful handling of connection issues and timeouts

Modern, interactive corporate website showcasing AI services, team profiles, blog content, and career opportunities

Industry/Domain

Technology / AI Services / Corp
This project delivers a corporate website for Vertex AI Tec designed to showcase AI services, team members, blog content, and career opportunities. The goal was to create a modern, interactive user experience using animations and WebGL components while maintaining a clean corporate identity.

The solution is a frontend-only React application featuring multi-page navigation, interactive UI components, and media-rich sections, built with performance-conscious animations using GSAP and lightweight WebGL effects.

Course Craft is an AI-powered educational tool that automates:

CLO–PLO Mapping (Course Learning Outcomes → Program Learning Outcomes)

Bloom’s Taxonomy Classification

Student Achievement Analysis

Built with Python, NLP, and Neo4j, Course Craft reduces course design time from days to minutes, ensuring accuracy, consistency, and accreditation readiness for educators and academic institutions.

🚀 Features

Automated CLO Generation – Extracts CLOs from course descriptions and lecture outlines.

PLO Mapping Engine – Uses NLP to match CLOs with relevant PLOs based on curriculum standards.

Bloom’s Taxonomy Classification – Categorizes CLOs into cognitive levels (Remember → Create).

Dynamic Achievement Calculation – Supports variable student datasets for CLO success rate analysis.

Interactive Web Interface – Easy to use via Flask or Streamlit.

Data Visualization – Displays CLO–PLO mappings and achievement rates in clear tables & charts.

🛠️ Tech Stack

Programming Language: Python

Frameworks: Flask / Streamlit

NLP Tools: HuggingFace Transformers, NLTK, BERT

Database: Neo4j (for relationship mapping)

Data Processing: Pandas, NumPy

Visualization: Matplotlib, Plotly