My expertise and research are rooted in machine learning for computer vision, with a recent focus on ensuring the safety of Generative AI (GenAI) technologies in visual contexts. I am dedicated to developing a comprehensive filtering system that addresses safety concerns in GenAI, targeting the mitigation of inappropriate, hateful, and privacy issue content. Before embarking on this path, my efforts were concentrated on Multimodal AI projects. These initiatives ranged from providing solutions for various image editing tasks—including inpainting, object removal, background alteration, and creating personalized content—to enhancing open-vocabulary object detection overcoming the constraints of predefined object categories through effective image-text alignment.