This study investigated the effectiveness of various models in deep learning in performing sentiment analysis on code-mixed Hinglish text, a hybrid language widely used in digital communication. Hinglish presents unique challenges due to its informal nature, frequent code-switching, and complex linguistic structure. This research leverages datasets from the SemEval-2020 Task 9 competition and employs models such as RNN (LSTM), BERT-LSTM, CNN, and a proposed Hybrid LSTM model along with CNN. The study’s primary objective is to develop a robust sentiment analysis framework that accurately classifies sentiment in Hinglish text. The Hybrid LSTM-CNN model, combining the strengths of LSTM and GRU units, demonstrated superior performance with an accuracy 93.76%, precision of 93.49%, and recall of 93.56%, with F1 of 93.54%. This model outperformed existing approaches, including the BERT model from the HinGE dataset, which achieved an accuracy of 76.18%. The results highlight the proposed model’s capability to handle the nuances of Hinglish, including its informal and code-mixed nature, more effectively than traditional models. The model also snags for future developments like data bias, interpretability of the model, scalability. This study establishes a underscores the significance in techniques in advanced deep learning in tackling the challenges of code-mixed language processing.