MLMD is an AI platform designed to predict and design novel materials without requiring programming skills. It integrates data analysis, descriptor refactoring, hyperparameter optimization, and property prediction, offering a user-friendly web-based interface. The platform enables end-to-end materials discovery, utilizing model inference, surrogate optimization, and active learning to identify materials with desired properties. MLMD is particularly effective in scenarios with limited data and supports various material types, including perovskites, steel, and high-entropy alloys. It enhances material discovery by integrating experimental and computational methods, accelerating the development of new materials. MLMD also includes tools for classification, regression, surrogate optimization, and active learning, demonstrating strong performance in predicting material properties and discovering novel materials. The platform's user-friendly interface and code-free operation make it accessible to researchers without programming experience, significantly advancing materials informatics. MLMD's ability to handle data scarcity and provide efficient material design solutions makes it a valuable tool for materials scientists. The platform's integration of AI techniques with materials science enables the discovery of materials with enhanced properties, such as high hardness and improved mechanical performance. MLMD's effectiveness is demonstrated through various case studies, including the design of RAFM steels and high-performance alloys. The platform's robustness and versatility make it a promising tool for accelerating materials research and development.MLMD is an AI platform designed to predict and design novel materials without requiring programming skills. It integrates data analysis, descriptor refactoring, hyperparameter optimization, and property prediction, offering a user-friendly web-based interface. The platform enables end-to-end materials discovery, utilizing model inference, surrogate optimization, and active learning to identify materials with desired properties. MLMD is particularly effective in scenarios with limited data and supports various material types, including perovskites, steel, and high-entropy alloys. It enhances material discovery by integrating experimental and computational methods, accelerating the development of new materials. MLMD also includes tools for classification, regression, surrogate optimization, and active learning, demonstrating strong performance in predicting material properties and discovering novel materials. The platform's user-friendly interface and code-free operation make it accessible to researchers without programming experience, significantly advancing materials informatics. MLMD's ability to handle data scarcity and provide efficient material design solutions makes it a valuable tool for materials scientists. The platform's integration of AI techniques with materials science enables the discovery of materials with enhanced properties, such as high hardness and improved mechanical performance. MLMD's effectiveness is demonstrated through various case studies, including the design of RAFM steels and high-performance alloys. The platform's robustness and versatility make it a promising tool for accelerating materials research and development.