Accurately distinguishing true hardware failures from false alarms is a critical requirement in large-scale optical networks, where unnecessary Return Material Authorizations (RMAs) result in significant operational and financial overhead. This paper presents a novel AI-driven predictive framework that integrates multi-domain telemetry fusion, Transformer-based temporal modeling, and a domain-aware hybrid ensemble to deliver carrier-grade hardware failure detection in optical embedded systems. Unlike prior works that rely on single-sensor or threshold-based diagnostics, the proposed approach jointly analyzes optical power fluctuations, laser bias-current drift, TEC thermal instability, voltage dynamics, and DSP-layer soft metrics, enabling the model to capture degradation signatures that emerge only through cross-sensor interactions. A customized ensemble combining Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN)-LSTM, and TimeSeriesBERT is introduced to fuse complementary pattern-recognition capabilities--including long-term drift modeling, high-frequency anomaly detection, and global multi-sensor attention--resulting in superior robustness and generalization. Evaluation of real-time telemetry from optical devices demonstrates the effectiveness of the proposed system, achieving high accuracy with a high F1-score and significantly reducing unnecessary RMAs. These results highlight the novelty and practical value of the presented framework, establishing it as the first comprehensive AI solution tailored for reliable hardware-failure prediction in optical embedded systems.

With the increasing complexity and miniaturization of printed circuit boards (PCBs), conventional inspection techniques are no longer sufficient to ensure high manufacturing yield and reliability. This paper presents an intelligent computer vision and machine learning–based framework for automated detection of hardware trace defects, soldering faults, missing components, and layout–schematic mismatches in PCB assemblies. The proposed system performs schematic-aware inspection by aligning captured PCB images with corresponding design netlists and layout data, enabling precise verification of routing continuity, component placement, and interconnect integrity. Advanced image preprocessing, feature extraction, and deep learning–based object detection and classification models are employed to identify subtle manufacturing defects with high accuracy. Experimental results on real PCB datasets demonstrate a significant improvement in defect-detection accuracy and a substantial reduction in false positives compared to conventional rule-based inspection methods. The proposed approach enables fast, non-contact, and scalable PCB inspection, making it highly suitable for next-generation electronics manufacturing, automated quality assurance, and predictive fault analysis.

The manuscript is accepted. It presents Kalya Research as a practical, well-scoped virtual assistant for CAM and non-pharmacological interventions, pairing curated lexicons with rule-based retrieval and clear visual analytics. The alopecia case study...