• Architected a computer vision solution to identify under-extrusion errors in 3D printing processes, integrating a fine-tuned ResNet-50 model with LDS edge detection to enhance spatial feature extraction. Tailored for high-throughput analysis and processed through rigorous validation and testing protocols to refine accuracy. • Achieved a precision of 75%, ranking 7th on a competitive leaderboard, showcasing the model's capability in real-time anomaly detection and potential for deployment in industrial quality control systems.