The growing prevalence of mental health challenges, such as anxiety and depression among students, highlights the urgent need for accessible early intervention. The essence of this project lies in the development of an easy to use online platform of personalized mental health support which utilizes sentiment analysis for recognizing early signs of distress. The system utilizes a structured questionnaire combined with advanced machine learning algorithms to analyze an user responses to identify emotional trends. We base our model on neural network architecture for binary sentiment classification (based on Google News Swivel embeddings) and using it as a core of customized feedback mechanism. In this initiative, we are integrating mental health educational content as well as targeted self-care exercises and access to direct consultant with mental health professionals to increase mental health awareness, support active forms of self-care and make it easier to seek help without stigma. This is a platform that will ultimately support students as well as those who are young and are at risk, meaningfully and accessibly, in order to ensure good mental health for all.