Computer Engineering Student · Dual Degree

Mai
Hamed

Building at the intersection of code, data & electrical systems.

Ain Shams University University of East London Cairo, Egypt
Mai Hamed
Mai Hamed
Mai Hamed
Mai Hamed
scroll ↓

Engineering
solutions,
not just code.

Mai Hamed
Available for Internships 🚀

I'm a 3rd-year Computer Engineering student holding a dual degree from Ain Shams University (Cairo) and the University of East London. My work spans the full stack — from low-level embedded systems and hardware design to data pipelines, machine learning models, and production web apps.

I'm passionate about building things that actually work: clean architecture, measurable outcomes, and real-world impact. Whether that's a 96.8%-accuracy digit classifier or a restaurant ordering system that cut manual processing time by 40%.

Outside of code, I lead and coordinate volunteer efforts at the Faculty of Engineering — recognized among the top 5% of contributors with the Excellence Award.

3+ Years Coding
10+ Projects
C1 English (IELTS)
Top 5% Volunteer Award

What I work with

⟨/⟩

Programming

C++PythonC#Java JavaScriptSwiftSQLDart MATLABVHDLMIPS Assembly

Data & AI / ML

pandasNumPyPySpark scikit-learnCNNOpenCV Power BIMS FabricmatplotlibseabornPyTorchTensorFlow

Web & Mobile

ReactNode.jsExpress MongoDBHTML/CSSFlutter REST APIsMVC/MVVM

Embedded & Tools

FreeRTOSTiva-C TM4C123GH6PMArduino UARTPID ControlEEPROM GitVS CodeMS SQL ServerPyCharm

Things I've built

Machine Learning · Python
AI / ML Computer Vision Data Analysis

MNIST Image Classifier

Built 6 classifiers (KNN, SVM, Decision Tree, Naïve Bayes, Logistic Regression, CNN) from scratch with NumPy. Achieved 96.8% accuracy with CNN feature extraction (+5.3%), PCA, L1/L2 regularization, and 5-fold cross-validation. Ships with an interactive GUI for real-time digit drawing and probability bar charts.

PythonNumPyCNNscikit-learnPCA
Anomaly Detection · PatchCore
AI / ML Computer Vision

DefectVision — Industrial Anomaly Detection

Unsupervised defect detection system using PatchCore with a ResNet18 backbone — no defect labels needed during training. Achieved 92.1% ROC-AUC and 88% accuracy on the MVTec bottle dataset. Optimized coreset memory bank to run on CPU with ~5x less RAM. Deployed as a Streamlit app with real-time anomaly heatmaps and a ROC-calibrated threshold.

PythonPyTorchPatchCore scikit-learnStreamlit
Data Mining · Classification
AI / ML Data Analysis

Mushroom Classification

Cleaned an 8,124-row categorical dataset with mode imputation & label encoding. Applied K-Modes clustering and Cramér's V / Theil's U correlation analysis. Compared Decision Tree, KNN, Random Forest, and Naïve Bayes — achieving 100% accuracy with tree-based models. Visualized ROC/AUC, confusion matrices, and feature importance.

Pythonpandasscikit-learnK-ModesROC/AUC
Unsupervised ML · Clustering
AI / ML Data Analysis

Customer Segmentation Analysis

Unsupervised ML pipeline on 2,240-customer marketing dataset. Applied PCA (3 components) for dimensionality reduction, Elbow Method to determine optimal cluster count, and Agglomerative Clustering to identify 4 distinct customer segments varying by income, spending, promotions response, and family profile.

PythonPCAscikit-learnseabornAgglomerative Clustering
Data Analysis · Business Intelligence
Data Analysis

Stockout & Phantom Revenue Analysis

Analyzed real ASOS product catalog data to uncover lost revenue from stockouts. Parsed messy size/stock strings, calculated phantom revenue per product, and aggregated by brand. Produced a brand strategy scatter plot with quadrant segmentation to identify high-value restocking targets.

PythonpandasmatplotlibseabornBusiness Intelligence
Computer Vision · OpenCV
Computer Vision AI / ML

Jigsaw Puzzle Solver

Modular pipeline to reassemble 2×2/4×4/8×8 puzzles using multi-feature border analysis — LAB color, Sobel gradients, Laplacian edges — with feature standardization. Integrated adaptive preprocessing (CLAHE, gamma correction, noise-aware denoising) and generated heatmaps, grid-cut diagrams, and quality-score visualizations.

PythonOpenCVNumPyCLAHE
Full-Stack · Web
Full-Stack

Reddit Clone

Reddit-style platform with 10+ RESTful API endpoints, AI-driven content features, and a MongoDB schema spanning 5+ data models for scalable user and post management.

ReactNode.jsExpressMongoDB
Agile · Scrum · Full-Stack
Full-Stack

University Management System

Led frontend development across 4 Scrum sprints, building 30+ React pages with role-based dashboards for Admin, Student, Staff, Advisor, and Applicant roles. Includes a real-time messaging system, enrollment, course management, hall booking, and full sidebar navigation per role.

ReactNode.jsExpressSQLScrum
E-Commerce · Web
Full-Stack

Cozy Coffee — Café Shop

Responsive café website with 100% mobile compatibility using CSS Flexbox and vanilla JS. Supports ordering, login, sign-up, add-to-cart, and checkout flows with cross-browser consistency.

HTMLCSSJavaScript
FreeRTOS · Tiva-C TM4C123GH6PM
Embedded

Smart Parking Garage Gate

Advanced real-time system implementing multitasking, obstacle protection, safety handling, and inter-task communication. Designed with semaphores, mutexes, queues, and FSM logic on a Tiva-C microcontroller running FreeRTOS.

CFreeRTOSTiva-CUARTFSM
Dual MCU · Layered Architecture
Embedded

Door Locker Security System

Layered embedded software using dual TM4C123 microcontrollers with UART communication, EEPROM storage for persistent password data, and motor-driven lock control.

CTiva-CUARTEEPROMMotor Control
Control Systems · Arduino
Embedded

PID Position Control System

Hardware closed-loop feedback control system using Arduino and a geared DC motor. Designed and tuned a PID controller for accurate angular position control.

C++ArduinoPIDDC Motor
Mobile · Flutter
App Dev

Basmei Yumi — Cooking App

Full mobile cooking app built with Flutter using MVVM architecture. Features recipe browsing, step-by-step instructions, and a polished cross-platform UI for iOS and Android.

FlutterDartMVVM
Database · Desktop App
Full-Stack

Restaurant Ordering System

8-table SQL database with 15+ stored procedures, cutting manual order processing time by ~40%. C# Windows Forms UI integrated directly with SQL Server.

SQL ServerC#Windows Forms
Algorithms · C++
Algorithms

Algorithms & Complexity Analysis

Implemented 6 algorithmic tasks — Tromino Tiling, Knight's Tour, Tower of Hanoi, BFS Knight Swap, DP Lattice Covering — with full complexity analysis across 10+ approaches.

C++BFSDPDivide & Conquer
E-Commerce · Web
Full-Stack

Beauty E-Commerce

Full-featured online beauty store with product listings, cart management, and a clean responsive UI designed for a smooth shopping experience.

HTMLCSSJavaScript
Game · Python
Algorithms

Snake Game

Real-time game loop running at 60 FPS with auto-scaling difficulty across 5 speed levels.

Python

My journey

Jul – Aug 2025

Mobile Development Trainee

National Telecommunication Institute (NTI)

160+ hours of intensive Flutter training. Built 3 fully functional app prototypes using MVC and MVVM architecture within a 6-week program. Achieved 100% of technical milestones and earned an official NTI Certificate of Completion.

FlutterMVCMVVMDart
Oct 2023 – Present

B.Sc. Computer Engineering

Ain Shams University, Cairo

Relevant courses: Data Structures, Web Development, Databases, Software Engineering, Embedded Systems, Machine Learning (CSE382), Data Mining & BI (CSE387).

Oct 2023 – Present

B.Sc. Computer Engineering

University of East London, London

Dual degree program running parallel to Ain Shams University curriculum.

2023 – Present

Volunteer Member · Excellence Award

Faculty of Engineering, Ain Shams University

Led 15+ volunteers across 3 major campaigns. Coordinated 4+ faculty-wide events per semester (200+ attendees each). Recognized among the top 5% of contributors with the Excellence Award.

Certificates

IELTS C1 — Advanced English
British Council · Jul 2024
Mobile Development with Flutter
Sprints · Jul 2025
NTI Mobile Development Certificate
National Telecommunication Institute · Aug 2025
AI for Beginners
HP LIFE · Jun 2025

Let's connect

I'm actively looking for internship opportunities. Whether it's a question, a role, or just a hello — my inbox is open.