Computer Science student specializing in Robotics & AI. Building intelligent systems at the intersection of deep learning, computer vision, evolutionary algorithms, and embedded hardware.
I am a Computer Science student in the Robotics & AI Department at Helwan National University (2023 – 2027). My technical focus centers on bridging theoretical artificial intelligence with real-world engineering — spanning deep learning, computer vision, optimization algorithms, and embedded hardware automation.
With hands-on proficiency in Python, C, Java, MATLAB, Linux, microcontrollers, and sensors, I design end-to-end solutions: from neural network vision classifiers and clustering pipelines to IoT smart access systems and evolutionary VRP routing algorithms.
AI & Software Engineering
Directly aligned with curriculum, certified programs, and real engineering implementations.
All 8 production projects spanning Deep Learning, Computer Vision, Optimization Heuristics, IoT Hardware, and Security.
A Deep learning-based waste classification system capable of identifying six waste categories: plastic, paper/cardboard, metal, glass, organic waste, and e-waste. Designed, trained, and evaluated multiple computer vision models, including custom CNN architectures, transfer learning models, and Vision Transformer (ViT).
Developed an end-to-end machine learning classification pipeline to predict customer churn using demographic, service usage, and billing-related data. Performed data cleaning, missing value handling, categorical encoding, class imbalance handling, and leakage prevention. Trained and evaluated Logistic Regression and hyperparameter-tuned Random Forest models using accuracy, precision, recall, F1, and ROC-AUC.
Developed an end-to-end unsupervised pipeline segmenting customers based on purchasing behavior, demographics, discounts, returns, and transactions. Integrated relational datasets, performed feature engineering and scaling, then applied and compared K-Means and Agglomerative Hierarchical Clustering using Silhouette Score and Davies-Bouldin Index.
Built a supervised machine learning regression model using Python and Scikit-learn to predict vehicle prices from structured data. Implemented preprocessing pipelines with StandardScaler and One-Hot Encoding, then trained and evaluated Ridge Regression and Random Forest Regressors via MAE, RMSE, R² score, and runtime efficiency.
Built an end-to-end computer vision pipeline using Python and OpenCV to automatically detect, segment, count, and classify coins from images. Applied CLAHE, Gaussian blur, morphological cleanup, distance transform, watershed segmentation, connected components, and radius-based sizing to calculate total monetary value with batch processing.
An optimization-based desktop application for capacity-constrained multi-vehicle routing (VRP). The system compares Genetic Algorithm (GA) and Differential Evolution (DE) approaches to generate efficient delivery routes, featuring an interactive GUI for route visualization, convergence curves, and algorithm comparison.
Built a smart access control system using ESP32 and embedded C/C++ to manage secure door locking and remote monitoring. Integrated a 4x4 keypad, LCD display, servo motor, PIR motion sensor, ultrasonic sensor, buzzer, Wi-Fi, and Telegram bot commands for PIN authentication, lockout, and real-time security alerts.
Performed a structured penetration testing assessment on a vulnerable web application. Identified and exploited vulnerabilities including SQL Injection, XSS, Command Injection, and Privilege Escalation. Conducted network reconnaissance, enumeration, password cracking, and authored a comprehensive remediation security report.
Accredited artificial intelligence coursework, industrial internships, and specialized deep learning credentials.
Intensive program covering Neural Networks & Deep Learning (24h), Data Prep & Exploration (12h), Numerical Optimization, Linear Algebra, Probability & Statistics, Python, and practical hands-on project.
Demonstrated competence in foundational deep learning architectures, convolutional neural networks, computer vision classification, and GPU-accelerated model training.
Constructing multi-layer perceptrons, backpropagation mechanisms, loss optimization functions, and evaluating neural network architectures in PyTorch.
Tokenization, text normalization, vectorization, word embeddings, TF-IDF representations, and building text classification models in Python.
Filtering, thresholding, edge detection, contrast adjustments, morphological transformations, and automated feature extraction from visual media.
Applied convolution operations, pooling mechanisms, transfer learning with modern backbones, data augmentation techniques, and custom image classifier pipelines.
Practical summer training at the Ministry of Petroleum & Mineral Resources covering engineering workflows, organizational systems, and industrial technical operations.
Available for AI & Software Engineering internships, research initiatives, and intelligent automation projects. Reach out directly through any channel below.