This paper describes the software architecture of the STRATIFYHF platform, a cloud-based, AI-powered decision support system for early detection, risk stratification and progression prediction of heart failure in primary and secondary care. The system follows a microservices-based, multi-tiered design with modules for data management, machine learning-based diagnosis and risk stratification, voice-based early diagnosis, computational disease-progression modelling, and interactive visual analytics, all coordinated through a centralized workflow manager and exposed via a web-based clinician interface supporting multimodal data (ECG, echocardiography, cardiac MRI, voice). This work was presented at the 4th Serbian International Conference on Applied Artificial Intelligence (SICAAI 2025), Zlatibor, Serbia, and was carried out within the STRATIFYHF project.