Nour Hatem

AI Engineer & Data Scientist | ML · LLMs · Analytics

AI systems are designed, data pipelines are built, and intelligent solutions are shipped from research to production.

About Me.

I started coding to understand how machines think — and never stopped.

I am an AI Engineer focused on building intelligent systems that solve real-world problems. My journey in artificial intelligence spans deep learning, natural language processing, computer vision, and production-grade machine learning pipelines.

With a foundation in Computer Science and Artificial Intelligence from Helwan University, I combine rigorous theoretical knowledge with hands-on engineering. I thrive at the intersection of research and deployment — where models move from notebooks to scalable production systems.

My work is driven by curiosity and precision: every model I build is designed not just to perform, but to be reproducible, interpretable, and deployable.

B.Sc. Computer Science & Artificial Intelligence

Helwan University

Digital Egypt Pioneers Initiative (DEPI) — Machine Learning Track

National Telecommunication Institute (NTI) — Data Analysis & ML Specialist

ITI × NVIDIA Deep Learning Institute — Generative AI & Deep Learning

Huawei HCIA-AI — AI & Machine Learning Certification

Experience.

During my final university year, I pursued intensive specialised training across five programs — deliberately building depth in deep learning, production ML, generative AI, and data analysis.

March 2026 – Present

Data Analysis Trainee

National Telecommunication Institute (NTI)

Performing advanced data analysis using Python, SQL, and Power BI to derive actionable insights from large-scale datasets.
Building interactive dashboards and automated reporting pipelines for stakeholder presentations.
Applying statistical methods and exploratory data analysis to support data-driven decision making.

February 2026

Generative AI & Deep Learning Trainee

ITI × NVIDIA Deep Learning Institute

Built LLM-powered applications using prompt engineering and RAG architectures during an intensive NVIDIA DLI program.
Implemented deep learning models for image classification and NLP tasks using PyTorch and TensorFlow.
Earned NVIDIA DLI certifications in Deep Learning and LLM Application Development.

November 2025 – Present

Machine Learning Trainee

Digital Egypt Pioneers Initiative (DEPI)

Developing end-to-end machine learning pipelines covering data preprocessing, feature engineering, model training, and evaluation.
Implementing supervised and unsupervised learning algorithms with scikit-learn and XGBoost.
Working on real-world projects including disease prediction, trip duration forecasting, and classification systems.
Tracking experiments with MLflow and deploying models via Streamlit and FastAPI.

October 2025

AI & Machine Learning Trainee

Huawei HCIA-AI

Studied machine learning fundamentals, deep learning architectures, and AI application development through Huawei's HCIA-AI program.
Studied neural network architectures including CNNs, RNNs, and attention mechanisms.
Applied learned concepts through lab exercises on Huawei's ModelArts platform, covering model training and deployment workflows.

July 2025 – August 2025

Machine Learning Specialist Trainee

National Telecommunication Institute (NTI)

Trained across regression, classification, clustering, and dimensionality reduction techniques on real-world datasets.
Built predictive models using scikit-learn, XGBoost, and ensemble methods on real-world datasets.
Recognised as top performer among cohort peers for project quality and delivery.

Projects.

AI & ML

Skin Cancer Detection

GitHub

EfficientNetV2-S transfer learning model on HAM10000 achieving 83.1% accuracy and 86.4% melanoma recall. Deployed as a Dockerized two-service FastAPI microservice.

AI & ML

NYC Taxi Trip Duration Pipeline

GitHub Kaggle

Production ML pipeline on 1.46M NYC taxi records using XGBoost inside a single sklearn Pipeline. Achieves R² of ~0.80 and RMSE of ~260 seconds.

AI & ML

Heart Disease Prediction System

GitHub Live Kaggle

Full ML pipeline with PCA preprocessing and XGBoost achieving 86.6% accuracy and 94.8% recall. Deployed as a live Streamlit app for real-time risk assessment.

AI & ML

Alzheimer's Disease Classification

GitHub Kaggle

End-to-end supervised ML pipeline on 2,149 clinical records benchmarking 10 models. XGBoost achieved 94.8% accuracy and 92.6% F1-Score, outperforming all baselines.