Ahmed Dridi

AI & Computer Vision Engineer

StartUp Names • Vollzeit

Jan 2026 - Apr 2026 • 3 mos

At ITGATE, I developed ShopVision, an end-to-end computer vision system for real-time visual product detection with integrated conversational AI. Trained YOLOv8 object detection model on 1,000+ curated product images achieving 68% mAP on 4-category product classification. Optimized inference pipeline from ~130ms per frame on CPU with 40% latency reduction through postprocessing optimization and model quantization. Implemented temporal filtering (multi-frame validation over 10-frame window) reducing false positives by 65% in real-world dynamic conditions. Architected PostgreSQL schema via Supabase for product metadata, detection logs, and search analytics with sub-200ms query response time. Shipped full-stack Streamlit application with product search UI, cart simulation, and real-time analytics dashboard deployed to production. Skills: Computer Vision, YOLO, Object Detection, TensorFlow, Streamlit, PostgreSQL, FastAPI, Docker