Hamail Shahbaz

I’m an AI engineer specialized in deep learning, computer vision, and data engineering, with hands-on experience implementing complex neural architectures for real-time semantic segmentation and medical imaging.

At AMENTOTECH, I built and refined production-ready backend systems—architecting scalable code, optimizing performance, and improving the Kyvo fintech app’s transactional efficiency and data processing speeds. I also supported the full software development lifecycle and integration testing as a Software Engineering Intern, collaborating on responsive user workflows and secure API integration for financial data management.

On the research and engineering side, I developed a real-time navigation system using BiSeNet with a ResNet-18 backbone, reaching 93.82% pixel-wise accuracy and adding speed-breaker detection logic that reduced linear velocity by 50% for smoother traversal. I also implemented U-Net and Attention U-Net for multi-region brain tumor segmentation on MRI scans, including data transformations and qualitative analysis using the BraTS 2020 dataset.

I’m especially driven by end-to-end MLOps: I engineered an automated pipeline with data ingestion, Great Expectations validation, and XGBoost training using a champion/challenger model registry and drift monitoring. I enjoy building reliable systems with strong engineering fundamentals—Python, PyTorch/TensorFlow, OpenCV, and scalable orchestration tools like Prefect—so models perform consistently in real-world conditions.