Ensuring an efficient and fair admission process is crucial for engineering colleges, as they need to predict the number of students enrolling in their courses accurately. Traditional approaches, such as manual calculations and fixed cut-off marks, often fail to account for dynamic factors like changing admission policies, reservation quotas, and variations in student performance over time. These outdated methods can lead to inefficient seat allocation, causing uncertainty amongst students and administrative challenges for colleges, particularly in India, where competition for engineering seats is intense. To enhance the accuracy of admission predictions, this study utilizes machine learning techniques, which can analyse historical data and identify trends that traditional methods overlook. We leverage admission data from 2020 to 2022 and apply advanced machine learning models, including XGBoost, support vector machine (SVM), and linear regression (LR). Prior to model training, we pre-process the data using feature engineering and data cleaning techniques to improve prediction accuracy. The models are evaluated using key performance metrics such as mean squared error (MSE), coefficient of determination (R2), and classification metrics like precision, recall, and F1-score. Our findings indicate that XGBoost and support vector machine (SVM) outperform linear regression, with XGBoost achieving an error score of 450 and SVM scoring 400, demonstrating their superior predictive capabilities. This research highlights the importance and significance of machine learning in optimizing admission forecasting, enabling colleges to allocate seats more efficiently whilst reducing uncertainty for students. Furthermore, by integrating additional factors such as student preferences, socio-economic background, and evolving education policies, future models can further refine their predictive power. The invention of this study contributes to the development of a data-driven, automated admission system that can streamline college admissions and support both institutions and students in making informed decisions.