B.Sc. Artificial Intelligence Engineering
Bahçeşehir University  ·  İstanbul, Türkiye

I’m an Artificial Intelligence Engineering student with an academic background across the UAE and Türkiye and hands-on experience building practical AI systems. My work spans machine learning, deep learning, computer vision, embedded systems, and cloud technologies, with a focus on designing efficient solutions that perform reliably under real-world constraints. I learn new technologies quickly and work effectively both independently and as part of a team.

• Education — B.Sc. in Artificial Intelligence Engineering at Bahçeşehir University, İstanbul, Türkiye (Expected 2027), following two years of Computer Engineering (AI Concentration) at Abu Dhabi University, Abu Dhabi, UAE.

• Focus Areas — Machine Learning, Deep Learning, and Computer Vision, with an emphasis on building practical and efficient AI systems and solutions.

• Selected Work — Developed an anatomy-aware lightweight CNN for retinal OCT disease classification achieving 99.90% test accuracy with approximately 8× fewer parameters than ResNet-50; built a smart face-recognition access-control system using Arduino with decisions returned within 6 seconds; and developed a cloud-native bookstore API using Flask and MongoDB, containerized with Docker and deployed on Kubernetes.

• Languages — Arabic (Native) · English (Fluent) · Turkish (Elementary Proficiency)

Languages

Deep Learning & Machine Learning

Deployment & Web

Databases & Hardware

Environments & Tools

Technologies: Python, PyTorch, OpenCV, NumPy, Kaggle (Tesla T4 GPU)

• Designed a lightweight CNN using self-supervised pretraining and anatomy-aware features.
• Introduced vertical position encoding and region-specialist branches to better model retinal anatomy.
• Achieved 99.90% test accuracy, matching ResNet-50 (99.79%) with 8× fewer parameters and 3× faster training.

Technologies: Arduino, C++, Python, OpenCV (LBPH), Proteus

• Built a smart access control system integrating Arduino with real-time face recognition.
• Implemented OpenCV's LBPH algorithm with NFC and IR remote as backup authentication.
• Detected users within 15 cm and returned a decision within 6 seconds for 5 enrolled users.

Technologies: Python, Flask, MongoDB, Docker, Kubernetes, Postman

• Built a cloud-native bookstore app exposing a RESTful API with Flask and MongoDB.
• Containerized with Docker and deployed on Kubernetes using StatefulSets and Services.
• Verified full CRUD functionality via Postman with persistent, in-cluster communication.

• walid.halabi@yahoo.com
• linkedin.com/in/walidmwkobahalabi