class MohamedWael:
role = "AI Engineer"
education = "B.Sc. AI & Data Science @ Zewail City of Science and Technology"
focus = ["Computer Vision", "Object Detection & Tracking", "RAG / LLM Apps", "MLOps"]
experience = ["AI Trainer & Prompt Engineer @ Outlier", "Freelance Software Tester @ test IO"]
ships = "end-to-end: dataset → model → API → dashboard"
def say_hi(self):
return "Let's build something that sees, reads and thinks."AI / ML
Backend · Frontend · Mobile · Ops
| Project | What it does | Highlights |
|---|---|---|
| 🎥 VisionOps | AI-powered retail CCTV analytics: people detection, tracking + re-identification, entry/exit counting, zones & dwell time, review dashboard | 93.2% mAP@50 · 92.3% precision (test set) · 138 → 46 track IDs with ReID · YOLO11s + ByteTrack + OSNet + FastAPI + React |
| 📄 SmartDoc RAG | RAG assistant for SEC 10-K filings with grounded, cited answers | FastAPI · FAISS · SentenceTransformers · Next.js |
| 🎳 Pin-Tracker | Fully offline bowling pin tracking & scoring app | 96.7% mAP@50 (standing pins) · 89.5% overall · YOLOv8n → TFLite · Flutter |
| ⚽ Football Analytics | End-to-end match analysis: players, ball & team assignment | YOLOv11 + BotSort |
| 🫁 Chest X-ray AI | Pneumonia detection from chest X-rays with confidence & report | Transfer learning · VGG16 · Deep Learning |
