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

Block or report mdhabibi

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
mdhabibi/README.md

Mahdi Habibi, Ph.D.

Data Scientist · AI / ML Engineer · Computational Physicist

Lead AI developer at DealCircle GmbH · Open-source author (TelescopeML) · Explainable AI (XAI) specialist

LinkedIn Portfolio Kaggle Medium Twitter


About

I am a senior data scientist and machine learning engineer with a Ph.D. in computational physics and more than 10 years of experience turning data into reliable, production-grade systems. My work spans the full lifecycle of AI/ML projects, from research and prototyping to deployment, in close collaboration with cross-functional teams.

  • 🔭 Lead developer of recommendation systems and AI applications at DealCircle GmbH.
  • 🧠 Specialist in Explainable AI (XAI), improving the transparency and trustworthiness of black-box models.
  • 📦 Author and maintainer of the open-source Python package TelescopeML, including its XAI module.
  • 🚀 Builder of end-to-end AI / ML / Deep Learning projects across computer vision, generative models, and time-series.

Focus Areas

Explainable AI (XAI) · Deep Learning · Computer Vision · Generative Models · Recommendation Systems · Time-Series Analysis · MLOps & Deployment


Skills & Tools

Languages

Python MATLAB Bash SQL HTML5

ML & Deep Learning

PyTorch TensorFlow Keras scikit-learn OpenCV

Data & Visualization

pandas NumPy Seaborn MySQL SQLite

Backend & Deployment

FastAPI Flask Django Docker AWS Azure Git Linux


GitHub Stats

GitHub stats Top languages


🏅 Certifications & Credentials

IBM Data Science Professional Certificate OpenCV University Johns Hopkins University DataScientest DeepLearning.AI


📂 Featured Projects

Project Techniques Data Types Poster
SOLID Design Principles Tutorial
Learn SOLID design principles through interactive Jupyter notebook tutorials.
SOLID principles, design patterns, OOP, software architecture, code quality SOLID poster
TelescopeML — Open-Source Python Package
Deep CNNs and ML models for analyzing stellar and exoplanetary telescope spectra.
Deep CNN, Machine Learning, XAI, Bayesian optimization, Feature Engineering Timeseries, Tabular TelescopeML poster
Malaria Cell Classifier
Deep CNNs and ML models for anomaly detection in microscopic malaria cells.
Deep CNN, Data Augmentation, Feature Engineering, Image Processing, Optimization Image Malaria classifier poster
BloodPy — Automated Blood Cell Classifier
Multi-classification of peripheral blood cells using deep CNNs and ML models.
Deep CNN, Data Augmentation, Transfer Learning, U-Net, Image Processing, Statistical Analysis, OpenCV, Fine-tuning Image, Metadata BloodPy poster
Dataset: Segmented Peripheral Blood Cells (OpenCV)
A dataset of segmented white blood cell images using advanced image processing.
GrabCut, Morphological Operations, OpenCV Image, Binary Masks, Dataset Blood cell dataset poster
LIME for Macroscopic Medical Images
A surrogate model (LIME) for enhancing transparency of medical diagnostics.
Deep CNN, LIME, XAI, Computer Vision, Optimization Image LIME poster
CAM for Macroscopic Medical Images
Class Activation Mapping (CAM) for anomaly localization interpretability.
Deep CNN, CAM, XAI, Computer Vision, Data Analysis, Optimization Image CAM poster
Automated Nucleus Detector
A semantic segmentation solution for automating nucleus detection in biomedical images.
U-Net, Keras-Tuner, Semantic Segmentation Image Nucleus detector poster
FastAPI Questionnaire API
A FastAPI application to manage and retrieve questionnaire data with authentication and custom error handling.
FastAPI, Authentication, Data Management, Error Handling, Shell API, CSV FastAPI poster
Neural Compression
Advanced autoencoder architecture for efficient lossless data compression.
Autoencoder, GenAI, SSIM, PSNR Image Neural compression poster
LIME for ECG Classification
A surrogate model (LIME) for enhancing transparency of time-series classification.
CNN, LIME, Time Series Analysis Timeseries ECG LIME poster
Beta-Variational Autoencoders
Generative learning (GenAI) with Beta-Variational Autoencoders.
Beta-VAEs, Latent Space Analysis Image Beta-VAE poster
Movie Recommendation Systems
Collaborative filtering, content-based filtering, and Bayesian average techniques.
Collaborative Filtering, Content-Based Filtering, Bayesian Average Metadata, User Ratings, CSV Recommender poster
Falcon 9 Rocket Predictor
A data-driven project to predict the success of Falcon 9 rocket landings.
Data Wrangling, Feature Engineering, Web-Scraping, JSON Processing, SQL, Hadoop, Folium, Decision Trees Unstructured, Tabular, Database Falcon 9 poster
Variational Autoencoders
A collection of VAE architectures built with the Keras deep learning framework.
Variational Autoencoders, Exploratory Data Analysis Image VAE poster
Netflix Content Analysis
Exploratory analysis of Netflix's catalog to uncover trends in content distribution and quality.
Exploratory Data Analysis, Data Visualization, Statistical Analysis Metadata Netflix analysis poster
Air Passenger Time-Series Analysis
Exploratory analysis of the Air Passengers dataset, uncovering trends and patterns over time.
Time-Series Analysis, Log Transformation, Moving Averages, Seasonal Decomposition Timeseries Air passenger poster

Open to collaboration on AI, machine learning, and explainable-AI projects. Feel free to reach out.

Popular repositories Loading

  1. LIME-for-Time-Series LIME-for-Time-Series Public template

    LIME for TimeSeries enhances AI transparency by providing LIME-based interpretability tools for time series models. It offers insights into model predictions, fostering trust and understanding in c…

    Jupyter Notebook 15 3

  2. CNN-Predictor-for-Malaria_Cells-LIME-CAM CNN-Predictor-for-Malaria_Cells-LIME-CAM Public

    Enhanced CNN model for malaria cell classification, featuring Class Activation Mapping (CAM) as a non-agnstic technique for anomaly localization and LIME (Local Interpretable-agnostic Explanation) …

    Jupyter Notebook 7 1

  3. DeepLearning-VAE DeepLearning-VAE Public

    Exploring the depths of generative learning with a $\beta$-Variational Autoencoder ($\beta$-VAE) applied to the MNIST dataset for robust digit reconstruction and latent space analysis.

    Jupyter Notebook 7 1

  4. Automated-Cell-Semantic-Segmentation-with-UNet Automated-Cell-Semantic-Segmentation-with-UNet Public

    A machine learning solution for automating nucleus detection in biomedical images, leveraging the U-Net architecture to accelerate medical research and disease treatment discovery.

    Jupyter Notebook 5 1

  5. Neural-Compression-with-Autoencoders Neural-Compression-with-Autoencoders Public

    Exploring advanced autoencoder architectures for efficient data compression on EMNIST dataset, focusing on high-fidelity image reconstruction with minimal information loss. This project tests vario…

    Jupyter Notebook 3

  6. Capstone-Project-IBM Capstone-Project-IBM Public

    A data-driven project to predict the success of Falcon 9 rocket landings, crucial for cost analysis and competitive strategy in the space industry. Involves data manipulation in Pandas, JSON data p…

    Jupyter Notebook 1