Computer Engineering Graduate • American University of Sharjah (AUS)

---

🚀 What I Do

I design and build innovative, intelligent systems that blend robust backend logic, intuitive frontend experiences, and applied machine learning to deliver impactful, end-to-end solutions.

• 🔄 Full Software Development Lifecycle (SDLC): From requirements analysis and system design to implementation, testing, and deployment
• 🤖 AI System Integration & RAG Workflows: Build context-aware intelligent systems using LLM integration, prompt engineering, retrieval-augmented generation (RAG), grounding strategies, and LangChain-based orchestration.
• 💻 Full-Stack Development: Build scalable web applications using modern frameworks, RESTful APIs, and cloud technologies
• 📊 Machine Learning & Data Science: Perform model training, data preprocessing, and feature engineering to extract actionable insights

---

💼 Project Portfolio

Welcome to my GitHub! Here you'll find my project folders, each highlighting a chapter of my journey across software engineering, AI systems, and data science.

Click below to explore 👇

🗂️ Software Engineering Projects

🧠 COMPAIR | View Project →

A full-stack comparison web application built with React.js and FastAPI, featuring AI integration through LangChain and OpenAI API for intelligent, structured comparisons. It combines robust frontend and backend engineering, API design, testing, and prompt-driven AI reasoning into a cohesive, reliable system.

⚙️ Automated Laundromat System | View Project →

Software Engineering SDLC project covering requirements (FRs & NFRs), design and analysis (UML diagrams: Use Case, Class, Domain Model, and Sequence), full-stack implementation (Firebase backend database and hosting + Flutter frontend), testing (plans and cases), and deployment of a smart laundromat web system.

🚗 Smart Toll Gate System | View Project →

An automated Smart Toll Gate System built using Raspberry Pi, integrating sensors and RFID for vehicle detection and identification, along with speed and weight measurement for dynamic toll calculation. The system controls gate access, uploads data to ThingSpeak for cloud monitoring, and supports operator overrides and alert mechanisms. It also includes RESTful Flask APIs that enable web-based gate operations, real-time vehicle information, and trespassing event handling through defined routes.

⚔️ Multithreaded RPG Engine with Design Patterns & Sensor Integration | View Project →

Java-based multithreaded RPG engine showcasing professional software architecture with advanced objected oriented programming. Implements six software design patterns Observer, Singleton, Template Method, Strategy, State, and Command. Uses concurrent threads for battle timing and real-time gyroscope, accelerometer, and sound input with synchronized, thread-safe execution.

🌤️ AI & IoT Based Air Quality Monitoring and Prediction | View Project →

Full-stack AIoT engineering project integrating embedded hardware, cloud infrastructure, and machine learning for real-time air quality monitoring and forecasting. Developed solar-powered ESP32 sensor nodes, a Firebase–GCP cloud pipeline, and a Flutter dashboard for live data visualization and predictive analytics using a hybrid CNN–BiLSTM model.

🗂️ AI & Machine Learning Projects

🧠 COMPAIR – AI-Powered Comparison Web Application | View Project →

A full-stack AI comparison web application built with FastAPI, React.js, LangChain, and GPT-4o, enhanced with Brave Search grounding for accurate, real-time context. Supports personalized, category-aware comparisons, schema-validated outputs, caching, and PostgreSQL-backed history with follow-up reasoning and shareable comparison links.

🛡️ Real-Time Fall Detection with TinyML | View Project →

On-device fall detection system for elderly care, running entirely on an Arduino Nano 33 BLE Sense with no phone or cloud dependency. Designed a selective multi-stage inference cascade that resolves everyday activity with lightweight arithmetic and only wakes the neural network when a fall is likely, achieving 97.5% live accuracy while cutting per-event energy by 98.4% versus an always-on CNN. Built the full pipeline from IMU data collection and windowing to threshold derivation, logistic-regression gating, CNN training, and TensorFlow Lite deployment as a 5.86 KB model on a Cortex-M4 board.

🌊 Semi-Supervised Learning Framework for Flood Prediction | View Project →

Semi-supervised machine learning framework for datasets with limited labels, demonstrating applied AI/ML and data science skills in predictive modeling. Includes rigorous data preprocessing, selective pseudo-labeling, and a three-phase training workflow, achieving strong result of test F1 = 0.9986, surpassing published benchmarks

🧬 Using Machine Learning to Predict Protein Stability | View Project →

Developed a supervised learning framework to predict protein stability upon single-point mutations using thermodynamic data derived from trypsin and chymotrypsin experiments. Implemented KNN, SVM, Decision Tree, and Naïve Bayes models with a rigorous pipeline including outlier removal (Z-score), standardization, class rebalancing (hybrid over- and undersampling), feature selection (SelectKBest), and dimensionality reduction (PCA). Performed 10-fold cross-validation and hyperparameter tuning across models, achieving strong generalization and robust predictive accuracy. Evaluated performance using F1-score, precision, recall, confusion matrix, and AUC-ROC curves to ensure stability predictions were reliable across all mutation classes

---

🛠️ Tech Stack

%%{init: {'theme': 'base', 'themeVariables': { 'primaryColor': '#6C63FF', 'primaryTextColor': '#ffffff', 'lineColor': '#8A2BE2', 'secondaryColor': '#b3a5ff'}}}%%
mindmap
  root((Tech Stack))
    Programming
      Python
      JavaScript
      Java
      C++
      SQL
      HTML
    Backend
      FastAPI
      Flask
      REST APIs
      OpenAI API
      Firebase API
      Google Maps API
    Frontend
      React.js
      Flutter
      HTML
      TailwindCSS
    AI/ML
      scikit-learn
      LangChain
      OpenAI API
      Feature Engineering
      Model Evaluation
      Prompt Engineering
      TensorFlow Lite
      TinyML
      Edge AI
    Data Science
      Matplotlib
      Seaborn
      Power BI
      Pandas
      NumPy
    Databases
      Oracle SQL
      Firebase Realtime
    Cloud
      Google Cloud
      Firebase Hosting
      Microsoft Azure

Loading

📋 Detailed Skills Breakdown

Programming Languages

Python Java C++ JavaScript SQL HTML

Backend Development

FastAPI Flask REST APIs OpenAI API Firebase API Google Maps API

Frontend Development

React.js Flutter HTML TailwindCSS

Artificial Intelligence & Machine Learning

• Frameworks: scikit-learn LangChain OpenAI API TensorFlow/Keras TensorFlow Lite
• Techniques: Supervised & Unsupervised Learning, Feature Engineering, Dimensionality Reduction (PCA), Model Evaluation (AUC, ROC), Hyperparameter Tuning, Prompt Engineering, TinyML & Edge AI Deployment, Model Quantization, CNNs

Data Science & Visualization

Matplotlib Seaborn Power BI

Databases

SQL (Oracle) NoSQL (Firebase Realtime Database)

Cloud Platforms

Google Cloud Platform (GCP) Firebase Hosting Microsoft Azure

---

💼 Current Focus

🔨 Enhancing COMPAIR — Expanding capabilities with improved grounding, structured output design, and cloud deployment

Combining my interests in full-stack engineering and AI systems engineering

---

🤝 Let's Connect!

Open to Collaborations & Opportunities

I’m always excited to connect around:

• Software Engineering and scalable system design
• Full-Stack Development using modern frameworks
• AI/ML-driven applications and intelligent automation
• Cloud-based architectures and deployment solutions
• Data Science projects focused on meaningful insights

---