In the backdrop of rapid industrialization and commercialization in developing nations like India, a grave threat to environmental sustainability is posed by the unchecked surge in fossil fuel consumption. With CO \(_{2}\) emissions reaching 1.58 metric tons as of 2020, the consequential impacts on the environment, human health, and biodiversity are increasingly evident. To safeguard life on Earth, Reducing and accurately predicting CO \(_{2}\) emissions are crucial steps that enable proactive measures to mitigate their impact. This project work is focused on predicting CO \(_{2}\) emissions in India until 2030, utilizing multivariate data encompassing various socioeconomic and environmental parameters. Deep learning methodologies, including Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and Temporal Convolutional Network (TCN), are employed to forecast emissions trends. Through rigorous evaluation, models that perform superiorly are selected for an ensemble approach. From the experimental results, it is observed that the LSTM and simple weighted average ensemble model exhibit better results. The ensemble model achieves an R-squared value of 0.86 for predicting CO \(_{2}\) emissions.