 |
 |
 |
 |
 |
|---|
- 🎓 Studying Information Systems at Science Valley Academy: a program built around software development, databases, and systems design
- 💼 ML Engineering Intern @ FlyRank AI, selected from 20,000+ applicants worldwide, completed all 11 core assignments plus a capstone project, Capstone Accepted.
- 🧑💻 Freelance For AI: certified through the ITIDA Gigs × eYouth freelance training program
- 🔬 AI Research: NIH Chest X-Ray multi-label classification preprint submitted to Research Square, now is live as a preprint on Research Square
- 📦 Open Source: deepcsv · 16,000+ downloads on PyPI
- 📈 Ranked #8 in Egypt for AI & ML Engineering creators | Favikon (June 2026)
- 🏆 15+ shipped projects in ML, medical AI, computer vision, and deep learning
- 🌱 Next up: exploring OCR and NLP
- 📫 Reach me at abdullah.bakr.official@gmail.com
- 🌍 Monufiya, Egypt
Information Systems: Science Valley Academy (High Institute for Engineering and Technology, Faculty of Science)
📅 [Sep 2025 – Jun 2029] | 🎯 Grade: Very Good
current_status = {
"internship": "ML Engineering Intern @ FlyRank AI (Jun 2026 – Sep 2026)",
"research": "NIH ChestX-Ray14 | Swin Transformer V2-B | 0.8495 mean AUC | is now live as a preprint on Research Square",
"open_source": "deepcsv | PyPI | 16,000+ downloads",
"learning": "OCR | NLP | etc",
"open_to": "AI/ML internships & full-time roles",
}| Project | Highlights | Stack |
|---|---|---|
| 🔬 NIH Chest X-Ray Classification 🔒 | 5 architectures benchmarked · 112K+ images · 0.8495 mean AUC · Swin V2-B surpasses CheXNet · Grad-CAM explainability · preprint under review at Research Square · (private — embargoed pending publication) | PyTorch, Swin Transformer, Grad-CAM |
| 🎭 Synthetic Image Attribution · ICANN 2026 | Dual-stream EfficientNet-B4 · RGB + SRM forensic features · 95%+ accuracy across 10 text-to-image models | PyTorch, EfficientNet |
| 🩺 Skin Lesion Segmentation | YOLO-based segmentation · melanoma vs. non-melanoma · trained on University of Waterloo VIP Lab dermatological datasets | PyTorch, YOLO |
| 🦷 Dental X-Ray AI | Three-stage pipeline · quadrant detection → tooth enumeration → pathology classification (caries, periapical lesions, impacted teeth) | YOLO, Object Detection |
| 🚁 Drone Detection in the Wild | YOLO26m + ResNet50 two-stage pipeline · 91.8% mAP@50 · 1,000+ real-world images | Ultralytics, Keras |
| 🎗️ Breast Cancer Classification | ResNet50 vs VGG16 vs EfficientNetB4 vs InceptionV3 vs MobileNet · 98.5% val acc · ultrasound images | TensorFlow, Keras |
| 😴 Drowsiness Detection | Real-time awake/sleepy classification · transfer learning on ~85K infrared eye images (MRL Eye Dataset) | PyTorch, Transfer Learning |
| 📦 deepcsv · PyPI | Auto data cleaning · 16,000+ downloads · GradientBoosting auto-feature-selection mode | Python, Pandas |
📂 More Projects
| Project | Highlights |
|---|---|
| FaceAge-Classifier | 5-class age group classification (18–60) · CNN & transfer learning on 69K+ face images |
| IBM Stock Forecasting | Time series forecasting (1980–2025) comparing Vanilla RNN, LSTM, and GRU |
| Stellar Object Classification | Multi-class classification (Galaxy/Star/Quasar) · stacked ensemble of LightGBM, XGBoost, ExtraTrees on SDSS17 data |
| Garbage Classification | 6-category classification · Inception V3 vs ResNet50 vs EfficientNet-B3 |
| Intel Image Classification | Multi-class CNN classification on the Intel Image dataset |
| Cats vs Dogs Classifier | Transfer learning · ~98% test acc · ResNet50 + augmentation |
| Weather Temperature Prediction | Gradient Boosting regression · R² = 0.687 |
| Iris Flower Classification | Classic ML pipeline comparing multiple classifiers with scikit-learn |
| FlyRank ML Internship Workspace | Notebooks, pipeline work, and capstone project from the FlyRank AI internship |
Deep Learning Frameworks
| TensorFlow | Keras | PyTorch |
|---|---|---|
 |
 |
 |
Computer Vision and ML
| OpenCV | Scikit-learn | NumPy | Pandas | SciPy |
|---|---|---|---|---|
 |
 |
 |
 |
 |
Specialized: Object Detection & Gradient Boosting
| Ultralytics | XGBoost | LightGBM |
|---|---|---|
 |
 |
.DjXN3gYZ.png) |
Visualization
| Matplotlib | Seaborn |
|---|---|
 |
 |
Tools and Environment
| Python | Jupyter | VS Code | Git | Kaggle |
|---|---|---|---|---|
 |
 |
 |
 |
 |
"I highly recommend Abdullah. He is a hardworking and consistent professional. We collaborated on a research paper together, and it was a pleasure working with him. He completed his part of the work within the given deadline. Abdullah has excellent programming skills in both Machine Learning and Deep Learning. I would recommend him for any AI-related project, particularly those involving Computer Vision or Large Language Models (LLMs)."
— Dr. Ashraf Ullah, Research Collaborator · July 2026
"I highly recommend Abdullah. We recently collaborated on a Kaggle competition where we joined in the final 10 days, and his dedication was key to our success. Abdullah is a brilliant problem solver who excels under pressure. He took the lead on building robust ML pipelines and fine-tuning hyperparameter spaces using Optuna, showing deep technical knowledge and agile thinking. He is a reliable teammate, an excellent communicator, and a true asset to any data science or AI team."
— Hozaifa Mohamed, Competition Collaborator · July 2026
| Language | Proficiency |
|---|---|
| 🇪🇬 Arabic | Native or bilingual proficiency |
| 🇬🇧 English | Professional working proficiency |
- 🧠 HCIA-AI V4.0 · Huawei ICT Academy (Aug 2026)
- 🤝 AI Ambassadors Program · National Telecommunication Institute (NTI) × Engineers for a Sustainable Egypt (Jul 2026)
- 💻 Freelancing Certification · ITIDA Gigs × eYouth (2026)
- 🚀 Build with AI: Masr Edition · Google for Developers (2026)
- 👁️ Deep Learning for Computer Vision · MaharaTech / ITI (2026)
- ☁️ AWS AI Practitioner Challenge · Udacity (2026)