We propose the utilization of Automated Machine Learning (AutoML) to automatically construct robust models for denoising strong gravitational lensing systems. AutoML harnesses advanced search and optimization techniques to autonomously explore, select, and fine-tune models, eliminating the need for manual hyperparameter tuning or specialized domain expertise. This approach offers a streamlined and impartial means of estimating model parameters efficiently, facilitating the development of versatile models applicable to diverse datasets without the need for repeated hyperparameter optimization. To assess the versatility of our AutoML model, we have gathered data from various astronomical instruments, including the Hubble Space Telescope (HST) and Kilo-Degree Survey (KiDS). Our findings indicate that AutoML not only facilitates generalized solutions for astronomical challenges but also excels in prediction accuracy, surpassing traditional approaches.