The study aims to develop an automated system leveraging modern machine learning techniques to identify defects in printed circuit boards (PCBs). Defects such as shorts, spurious copper, and missing holes can significantly compromise electronic devices’ reliability and performance. Traditional manual inspection methods are time-consuming and error prone. Thus, this work addresses these issues by using a custom PCB defect dataset annotated with Roboflow and training a YOLOv5 model on Google Colab. The system integrates an Arducam Camera Module IMX219 with a Raspberry Pi 4 to capture high-resolution PCB images, which are then inspected in real-time for defects. By combining advanced object detection algorithms with affordable hardware, this method offers a practical and cost-effective solution for PCB defect detection. Results demonstrated that the system is highly accurate and efficient in detecting PCB flaws, with performance metrics of 0.92, 0.89, and 0.90 for precision, recall, and F1 score, respectively. This significantly enhances quality control in PCB manufacturing.