Skip to content
View 0xNic11's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report 0xNic11

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
0xNic11/README.md

Abdullah Ashraf

Machine Learning Engineer · AI Practitioner · Cairo, Egypt

LinkedIn Twitter GitHub Email

Open to Work


About Me

I'm a passionate Machine Learning and AI practitioner with 2+ years of hands-on experience building end-to-end intelligent systems from raw data to deployed, monitored models.

My focus lies in crafting robust, production-ready ML pipelines with a strong emphasis on reproducibility, explainability, and real-world performance. I enjoy working across the full ML lifecycle: exploratory analysis, feature engineering, model selection, threshold tuning, and robustness evaluation.

  • 🔭 Currently building ML projects focused on fraud detection, credit risk, and churn prediction
  • 🌱 Deepening expertise in ensemble methods, MLOps, and model monitoring
  • 🧠 I once implemented gradient descent from scratch just to make sure I actually understood it. Turns out I didn't, then I did
  • ⚡ Firm believer that a good confusion matrix tells a better story than accuracy ever will
  • 📍 Based in Cairo, Egypt

Tech Stack

Languages

Python SQL

Machine Learning

scikit-learn XGBoost Random Forest TensorFlow PyTorch Keras

Data & Analysis

Pandas NumPy

Visualization

Matplotlib Seaborn

Environment & Tools

Jupyter Google Colab Git GitHub VSCode


Featured Projects

Project Description Tech
naked-ml Popular ML algorithms built from scratch using only Python built-ins and NumPy — for deep educational understanding Python NumPy
fraud-ml-deployment End-to-end fraud detection pipeline with threshold tuning, robustness analysis, and a full monitoring plan Python scikit-learn
credit-default-ensemble-ml Credit default prediction using Random Forest & Gradient Boosting with probability-based error analysis Python scikit-learn
customer-churn-ml Leakage-safe churn prediction pipeline with reproducible evaluation and baseline modeling Python scikit-learn
netflix-pandas-project Multi-table Pandas mini-project simulating a real-world data environment with programmatically generated auxiliary tables Python Pandas
medical-no-show-analysis EDA of medical no-shows with data cleaning, feature engineering, and visualization for ML readiness Python Pandas

GitHub Stats


Let's Connect

I'm actively looking for ML Engineering or Data Science roles where I can contribute to meaningful, real-world impact.

📬 Reach me at abdallahashraf5501@icloud.com, via LinkedIn, or on Twitter/X — always happy to connect.

"Build things that work. Understand why they work."

Pinned Loading

  1. naked-ml naked-ml Public

    Popular Machine Learning algorithms built from scratch using only Python Built-ins and NumPy. for educational purposes.

    Python 2

  2. fraud-ml-deployment fraud-ml-deployment Public

    End-to-end fraud detection ML project with threshold tuning, robustness analysis, and monitoring plan.

    Jupyter Notebook

  3. credit-default-ensemble-ml credit-default-ensemble-ml Public

    Credit default prediction using machine learning with Random Forest and Gradient Boosting, including probability-based error analysis and threshold sensitivity evaluation.

    Jupyter Notebook

  4. customer-churn-ml customer-churn-ml Public

    End-to-end machine learning project for customer churn prediction with leakage-safe preprocessing, baseline modeling, and reproducible evaluation.

    Jupyter Notebook

  5. netflix-pandas-project netflix-pandas-project Public

    Pandas mini-project using Netflix titles. Only one dataset is provided (`netflix_titles.csv`), so auxiliary tables (`credits`, `ratings`) are generated programmatically to simulate real multi-table…

    Jupyter Notebook

  6. medical-no-show-analysis medical-no-show-analysis Public

    Exploratory data analysis of medical appointment no-shows, focusing on data cleaning, feature engineering, and visualization to prepare an ML-ready dataset.

    Jupyter Notebook