Skip to content
View Abdo-ateM's full-sized avatar

Block or report Abdo-ateM

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 is supported. This note will only be visible to you.
Report abuse

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

Report abuse
Abdo-ateM/README.md

Hi, I'm Abdelrahman Hatem 👋

AI/ML student focused on Explainable AI, Machine Learning Engineering, MLOps, and Open-Source Python tools.

I am building my profile through real open-source contributions and practical AI/ML projects that connect model development, explainability, monitoring, and reproducible machine learning workflows.


Open-Source Contributions

SHAP — Explainable AI

I opened a bug-fix pull request to SHAP, a widely used explainable AI library.

Contribution: fixed Explanation.cohorts() failing with integer and boolean cohort labels by safely converting non-string cohort keys before passing them into Cohorts(**kwargs).

PR: shap/shap#4816

What this shows:

  • Python debugging and bug fixing
  • Understanding of SHAP internals
  • Regression testing
  • Contribution to real-world Explainable AI tooling

sktime-mcp — Agentic AI / MCP Tooling

I merged a documentation pull request to sktime-mcp, an MCP server that exposes sktime’s time-series machine learning registry and workflows to LLM agents.

Contribution: clarified first-time setup instructions, .venv usage, and MCP client fallback guidance using python -m sktime_mcp.server.

Status: Merged

PR: sktime/sktime-mcp#299

What this shows:

  • Merged open-source documentation contribution
  • Agentic AI / MCP ecosystem awareness
  • Developer tooling and setup clarity
  • Practical understanding of Python environments

Featured Projects

Model Explainability Dashboard

Interactive SHAP dashboard for tabular machine learning models using Streamlit, scikit-learn, and SHAP.

Focus areas:

  • Global feature importance
  • Local prediction explanations
  • Random Forest classification
  • Explainable AI dashboard design

Repository: https://github.com/Abdo-ateM/model-explainability-dashboard


ML Model Monitoring Dashboard

Lightweight ML monitoring dashboard for tracking model performance, prediction drift, and feature distribution changes.

Focus areas:

  • Model evaluation
  • Feature drift detection
  • Simulated production data
  • Streamlit monitoring dashboard
  • Practical MLOps concepts

Repository: https://github.com/Abdo-ateM/ml-model-monitoring-dashboard


Adult Income ML Pipeline

End-to-end machine learning pipeline for Adult Income classification using scikit-learn, ColumnTransformer, cross-validation, and model evaluation.

Focus areas:

  • Tabular data preprocessing
  • Classification modeling
  • Logistic Regression and Random Forest
  • Cross-validation
  • Saved evaluation artifacts

Repository: https://github.com/Abdo-ateM/adult-income-ml-pipeline


Machine Learning Regularization Comparison

Research-style comparison of regularization techniques in neural networks using TensorFlow/Keras and MNIST.

Techniques compared:

  • L1 Regularization
  • L2 Regularization
  • Dropout
  • Batch Normalization
  • Elastic Net

Repository: https://github.com/Abdo-ateM/machine-learning-regularization-comparison


Computer Vision / Video Colorization

Deep learning video colorization pipeline using MobileNetV2 U-Net, TensorFlow/Keras, OpenCV, LAB color space, and FFmpeg.

Focus areas:

  • Computer vision
  • Deep learning pipeline design
  • Frame-by-frame video processing
  • Image color space transformation

Repository: https://github.com/Abdo-ateM/video-colorization-deep-learning


Technical Skills

Languages: Python, C++, Java, SQL

Machine Learning: scikit-learn, TensorFlow/Keras, SHAP, model evaluation, feature engineering, neural networks

Data: pandas, NumPy, matplotlib, seaborn, preprocessing pipelines, cross-validation

MLOps / Engineering: Streamlit, Git, GitHub, virtual environments, reproducible project structure, model monitoring basics

Open Source: pull requests, issue discussion, regression tests, documentation improvement


Current Interests

  • Explainable AI and model interpretability
  • Open-source AI/ML tools
  • Agentic AI and MCP-based workflows
  • Practical machine learning engineering
  • Model monitoring and evaluation
  • Healthcare AI and real-world ML applications

Contact

GitHub: https://github.com/Abdo-ateM
Email: abdelgalilabdelrahman0@gmail.com

Pinned Loading

  1. shap shap Public

    Forked from shap/shap

    A game theoretic approach to explain the output of any machine learning model.

    Jupyter Notebook

  2. sktime-mcp sktime-mcp Public

    Forked from sktime/sktime-mcp

    Model Context Protocol server for exposing sktime’s time-series ML registry and workflows to LLM agents.

    Python

  3. model-explainability-dashboard model-explainability-dashboard Public

    Interactive SHAP explainability dashboard for tabular machine learning models using scikit-learn and Streamlit.

    Python

  4. ml-model-monitoring-dashboard ml-model-monitoring-dashboard Public

    Lightweight machine learning monitoring dashboard for tracking performance, prediction drift, and feature distribution changes.

    Python

  5. adult-income-ml-pipeline adult-income-ml-pipeline Public

    End-to-end machine learning pipeline for Adult Income classification using scikit-learn, ColumnTransformer, cross-validation, and model evaluation.

    Jupyter Notebook

  6. machine-learning-regularization-comparison machine-learning-regularization-comparison Public

    Comparative analysis of L1, L2, Dropout, Batch Normalization, and Elastic Net regularization in deep neural networks using TensorFlow and MNIST.

    Jupyter Notebook