Developed a Quantization Robustness Guard for the Samsung/ONE repository, focusing on enhancing the reliability of quantized model deployments. The solution introduced a pre-quantization validation step that detects and manages quantization-unfriendly values before they enter the quantization pipeline, thereby reducing the risk of misquantizations and runtime errors. This work involved compiler development and applied quantization techniques using C++. By addressing a critical bug rather than adding new features, the contribution improved the stability and performance of production systems relying on quantized models, ensuring that only suitable data proceeds through the quantization process and supporting more robust deployment outcomes.