UniPool — University Carpooling Platform
A mobile carpooling app matching university students by route and schedule. I designed the relational data model, the rider-driver matching logic (route proximity + schedule overlap), and the system architecture, and ran 20+ user interviews to validate demand. Co-founder and project lead across a 4-person team.
Stack: Flutter, Node.js, MySQL
0
3
A Python pipeline that pulls jobs from multiple sources, scores each listing against a profile using a weighted rule-based engine, and publishes a live ranked dashboard. Fully automated via GitHub Actions CI/CD — refreshes twice daily with no manual input. Demonstrates end-to-end delivery: data ingestion, scoring logic, static-site deployment, and scheduled cloud automation.
Stack: Python, GitHub Actions, HTML/CSS/JS, REST APIs
Link: github.com/MohanadddAmrr/job-engine
0
6
End-to-end computer vision pipeline that converts broadcast football video into 70+ tactical features. Led a 6-person team. A dual-detector system (YOLOv8 + RT-DETR second pass) pushed ball detection from 4.7% to 85.7%; ByteTrack handles identity continuity; jersey OCR resolves player identity. Runs at 25 FPS on consumer hardware across ~120K tracked frames. Delivered as a 16-tab Streamlit dashboard with xG modeling, formation detection, and pressing analysis.
Stack: Python, PyTorch, YOLOv8, RT-DETR, ByteTrack, OpenCV, Streamlit
Link: github.com/MohanadddAmrr/Tactivision
0
8
A 3-layer CNN (32–64–128 filters, dropout 0.5) for 7-class facial emotion classification on the CK+ dataset. Achieved 97% test accuracy and a 0.97 weighted F1, benchmarked against KNN, SVM, MLP, and MobileNetV2. Full pipeline: preprocessing, training, evaluation, and model comparison.
Stack: Python, TensorFlow, OpenCV, scikit-learn
Link: github.com/MohanadddAmrr/Emotion-Recognition-CNN
0
12