With the increasing complexity and scale of additive manufacturing, the need for efficient real-time inspection to detect errors when 3D printing has become more pronounced. It is essential to identify faults as early as possible because not only does it help to ensure that the output product maintains high quality, but it also decreases material waste and production delays. This project presented a practical and responsive solution using the YOLOv5 object detection model, considered one of the fastest and most accurate object detection models, to identify faulty printing errors as they occur during printing. An ESP32-CAM module was used to stream live video of the printing process, and can then detect faults in the printing process (such as warping, stringing, misalignment of layers), and when it detects a fault, an alert message is sent out to the users via a Telegram bot. This enables the users to address the fault in real time, with as little disruption to their workflow as possible. By utilizing image detection with automated communication, the system makes a 3D production run process that much more reliable. The project enhanced the intelligence of the manufacturing process by providing a proactive quality assurance and maintenance tool, which relates to the broader principles of Industry 4.0.