Sexism in social networks manifests in various forms, from blatant misogyny to subtle, implicit biases, presenting a significant societal challenge that necessitates effective detection and mitigation strategies. Addressing this issue involves participation in the EXIST 2024 tasks, a competition designed to advance the identification of sexist content in social media. This year’s contest includes both traditional text-based data from tweets and an innovative meme dataset, incorporating both images and text. The approach leverages sophisticated models to analyze these multimodal inputs. For textual modalities, a Convolutional Neural Network – Bidirectional Long Short-Term Memory model is employed to discern sexist language and tweet behaviours. For image modalities, a combination of Residual Network 50 and text-based analysis is utilized to detect and interpret sexist elements within memes. Both models undergo hyperparameter tuning and k-fold cross-validation to ensure robustness and accuracy. Preliminary results indicate that integrating these methods enhances the precision and effectiveness of sexism detection, providing a comprehensive tool for identifying and addressing sexist content in diverse social media formats.