05/08/2026 updated
100 % available
Machine Learning Engineer | Python & Scikit-learn
Aqaba, Egypt
Only remote
B.Sc. in Information Technology (ongoing)About me
I’m Ziad, a Machine Learning Engineer and Data Analyst. I help clients clean, analyze, visualize data, and build machine learning models that turn raw datasets into clear insights and smart decisions.
Data AnalysisGitHubPython (Programming Language)Machine LearningAzure Machine LearningSQL DatabasesSupervised LearningPylon Synchronization SoftwareFeature EngineeringRandom ForestJupyterGitFastAPIKotlinScikit Learn
Machine Learning & Model Development
Hands-on experience building end-to-end machine learning pipelines using Scikit-learn, covering classification, regression, and imbalanced learning. Expertise in model evaluation, hyperparameter tuning with GridSearchCV, and feature engineering.
Python & Data Engineering
Proficiency in Python with extensive use of Pandas, NumPy, and SQL for data cleaning, exploratory data analysis, and data visualization. Strong foundation in preparing and transforming datasets for machine learning workflows.
Deployment & API Development
Experience deploying machine learning models through Streamlit web applications and FastAPI REST APIs. Utilization of Joblib for model serialization and serving predictions in production-ready environments.
Random Forest & Ensemble Methods
Application of Random Forest Regressor and Classifier models for tasks such as software project cost estimation and customer churn prediction, achieving high performance metrics including R squared of 0.96.
Imbalanced Learning
Specialized handling of highly imbalanced datasets using techniques evaluated through Precision, Recall, F1-score, and Confusion Matrix, achieving 96% precision in fraud detection scenarios.
Scikit-learn Pipelines
Construction of reusable end-to-end pipelines combining preprocessing, model training, and inference using Scikit-learn Pipeline, enabling clean and maintainable ML workflows.
Version Control & Development Tools
Familiarity with Git and GitHub for version control, alongside Jupyter Notebook, Google Colab, and VS Code as primary development environments.
Logistic Regression & Decision Trees
Knowledge and application of Logistic Regression and Decision Trees as core supervised learning algorithms for classification tasks.
Hands-on experience building end-to-end machine learning pipelines using Scikit-learn, covering classification, regression, and imbalanced learning. Expertise in model evaluation, hyperparameter tuning with GridSearchCV, and feature engineering.
Python & Data Engineering
Proficiency in Python with extensive use of Pandas, NumPy, and SQL for data cleaning, exploratory data analysis, and data visualization. Strong foundation in preparing and transforming datasets for machine learning workflows.
Deployment & API Development
Experience deploying machine learning models through Streamlit web applications and FastAPI REST APIs. Utilization of Joblib for model serialization and serving predictions in production-ready environments.
Random Forest & Ensemble Methods
Application of Random Forest Regressor and Classifier models for tasks such as software project cost estimation and customer churn prediction, achieving high performance metrics including R squared of 0.96.
Imbalanced Learning
Specialized handling of highly imbalanced datasets using techniques evaluated through Precision, Recall, F1-score, and Confusion Matrix, achieving 96% precision in fraud detection scenarios.
Scikit-learn Pipelines
Construction of reusable end-to-end pipelines combining preprocessing, model training, and inference using Scikit-learn Pipeline, enabling clean and maintainable ML workflows.
Version Control & Development Tools
Familiarity with Git and GitHub for version control, alongside Jupyter Notebook, Google Colab, and VS Code as primary development environments.
Logistic Regression & Decision Trees
Knowledge and application of Logistic Regression and Decision Trees as core supervised learning algorithms for classification tasks.
Languages
ArabicNative speakerEnglishFluent
Project history
Developed a regression model to estimate car prices based on vehicle features. Built a FastAPI application to serve model predictions through a REST API. Applied data preprocessing, model training, and evaluation to improve prediction reliability.
Built a machine learning pipeline to detect fraudulent credit card transactions in a highly imbalanced dataset with only 0.17% minority class. Evaluated model performance using Precision, Recall, F1-score, and Confusion Matrix. Achieved 96% precision and reduced false positives through model tuning and Precision-Recall trade-off analysis.
Built an end-to-end classification pipeline to predict customer churn and identify high-risk customers. Used Scikit-learn Pipeline to combine preprocessing, model training, and inference in a reusable workflow. Applied GridSearchCV for hyperparameter tuning and deployed the trained model through an interactive Streamlit application.