Worked on quantization and model optimization features for edge AI workflows, focusing on the google-ai-edge/LiteRT and google-ai-edge/ai-edge-quantizer repositories. Developed a configurable enhancement to the TFLite calibration and quantization pipeline, allowing users to selectively disable per-channel quantization for dense layers, which provides finer control over the trade-off between model accuracy and inference performance. Additionally, implemented end-to-end Mean Squared Error–based quantization materialization for convolutional layers, integrating it into the algorithm manager to streamline quantization workflows. Leveraged C++, Python, and TensorFlow Lite, demonstrating depth in AI development, quantization techniques, and practical model optimization for edge deployment scenarios.