About Me

A Computer Engineering graduate from Air University, Islamabad, with expertise in Computer Vision and API Development. Currently exploring deployment and AI to drive intelligent solutions.

burhan.ahmed60090@gmail.com

My Interests

Computer VisionMLOpsDeep LearningEdge Computing
Burhan

Muhammad Burhan Ahmed

Experience

Junior Data Scientist

OIRRC | November 2025 - Present

Contributing to data-driven R&D projects in healthcare, focusing on ocular imaging. Built and evaluated deep learning models and deployed services to support various tools and applications within the institute.

Lab Engineer

Air University | August 2025 - Present

Teaching and assisting students in lab sessions, supporting course experiments and hands-on projects.

AI Intern

Jantah Tech | July 2023 - September 2023

Developed chatbots using Flask and SQLite for local datasets, enabling features such as appointment booking and patient query handling. Gained practical experience with Cloudinary, Git for version control, and Postman for API testing.

Computer Vision Engineer & GenAI Intern

Botmer International | April 2025 - June 2025

Worked on an AI-powered application for visual scene understanding in industrial safety environments. Managed the end-to-end DL pipeline for real-time inference. Deployed models using FastAPI. Leveraged tools including Ultralytics YOLO, Roboflow, TensorBoard and OpenCV.

Research Work

Paper Title: SWiM3: Solid Waste Mapping Dataset

Published At: 5th International IEEE Conference on Digital Futures and Transformative Technologies | Status: Submitted/Done

Developed a solid waste dataset containing 6,429 images and 25,234 annotated instances across 3 categories.

Paper Title: An Explainable and Edge-Deployable Deep Learning Framework for Smart Waste Classification

Published At: Journal | Status: In Progress

Designed and deployed a deep learning-based smart waste classification system on a Raspberry Pi, enabling edge inference. Conducted a comparative study of state-of-the-art object detection models to identify the most effective architecture for real-time performance. Integrated explainable AI techniques to interpret model predictions.