Heatwaves have become increasingly recurrent and intense in the current years, highlighting the need for accurate prediction models to mitigate their adverse impacts. This study proposes a heatwave prediction approach utilizing support vector regression (SVR) and random forest (RF) algorithms based on meteorological variables. Rajasthan has been selected as the study area that is an arid region and is more susceptible to heatwaves. In this study, historical meteorological data which includes air temperature, relative humidity, geopotential height, u-wind, and v-wind speed from the ECMWF Reanalysis version 5 (ERA5) for the time period 1991–2020 are used as the predictors. The peak temperature of summer months (April, May, and June) is predicted and acquired from the Indian Meteorological Department (IMD) for the same time period, i.e., 1991–2020. Both SVR and RF algorithms are utilized to model the complex associations among the meteorological variables and the heatwave events. SVR utilizes a kernel-based approach to map the input variables into a higher dimensional feature space, while RF employs an ensemble of decision trees to capture non-linear interactions. Model hyperparameters are fine-tuned to optimize performance. The prediction models are evaluated using various performance metrics, such as root mean squared error (RMSE) and coefficient of correlation (R). Future work involves expanding the scope of the study to incorporate additional meteorological variables and exploring the integration of other machine learning algorithms. Furthermore, the models can be enhanced by incorporating data from remote sensing sources and considering the spatial aspects of heatwave prediction. Such advancements will contribute to more accurate and reliable heatwave forecasting systems, enabling better preparedness and resilience against extreme heat events.