Noor Mustafa Chohan

I’m an AI/ML and front-end developer with hands-on experience building computer-vision pipelines, LLM-based chatbots, and client-facing web products. I enjoy owning a feature end-to-end, from data pipeline to deployed interface, and I bring that same momentum to every project.

As my Final Year Project, I’m currently leading an Exam Proctoring Solution—a YOLO-based exam-integrity detection system. I’m unifying five heterogeneous datasets into a single four-class YOLO training set using Roboflow for class remapping and format conversion, and I’m integrating YOLOv8n detection, DeepSORT tracking, and MediaPipe FaceMesh with a custom scoring-based cheating logic layer.

In my AI/ML Developer Intern role at Arch Technologies, I delivered four end-to-end AI/ML projects across NLP, computer vision, and predictive analytics. I built a Resume Intelligence System (TF-IDF + Logistic Regression), a Stock Trend Prediction web app with Flask from live financial data, a real-time Handwritten Digit Recognition system (CNN with TensorFlow/Keras + Pygame interface), and a Brain Tumor Segmentation pipeline (YOLOv11 + SAM2) that reached 84% segmentation accuracy.

I’ve also built user-facing products through front-end experience, including front-end interfaces for real-time system dashboards as an intern at National Radio Telecommunication Corporation (NRTC). As a freelancer, I’ve created an AI Chat Application (React + OpenAI API), an e-commerce platform (React + Redux + Firebase), an Admin Dashboard (React + Chart.js), a Kanban Board (React + React DnD), and a Weather Dashboard (React + REST API).