As modern software systems grow in complexity and scale, traditional approaches to fault tolerance and resilience are no longer sufficient. In this talk, we explore a paradigm shift in building truly self-healing software using Multi-Agent Systems (MAS). Inspired by biological systems and collective intelligence, MAS enables decentralised, autonomous, and cooperative behaviour across software agents, making them well-suited for dynamic, unpredictable environments.

We'll begin by demystifying the core principles of Multi-Agent Systems and their alignment with self-healing capabilities such as fault detection, diagnosis, recovery, and adaptation. Using real-world examples and architectures, we’ll demonstrate how agents can monitor, reason, and act to remediate issues—often before human intervention is required. Attendees will gain insights into the design patterns, communication protocols, and coordination mechanisms essential for building robust agent-based systems.