MLSysOps aims to advance AI-driven system adaptation across the cloud-edge continuum by developing a comprehensive framework for autonomic system management and application deployment. Using a hierarchical agent-based AI architecture, the project enables continual learning, intelligent retraining, and flexible, efficient execution on heterogeneous infrastructures.

Key features include energy efficiency, resilience, low latency, explainable ML, and secure, trusted storage and orchestration. The framework integrates with existing control systems and is validated through testbeds in smart cities and agriculture, combining real-world data and simulations to ensure scalability and impact.