As a telecom operator, we are exposed to the challenges of maintaining a reliable and robust network amidst environmental challenges, natural disasters and increasing customer demands. To address them, we envision developing an AI system for telecom network management to revolutionize how we operate by seamlessly integrating fault, performance, and alarm management into a cohesive, automated framework. Fault Management: By correlating diverse data sources—from network logs to user complaints—our AI system will pinpoint the root causes of faults swiftly and accurately. Performance Management: Our AI system will continuously monitor key performance indicators (KPIs). Through machine learning algorithms, it will identify patterns, predict potential performance degradation, and recommend proactive measures to maintain optimal network efficiency. Alarm Management: Our AI system will intelligently prioritize alarms based on their severity and potential impact on service quality. By correlating alarms with ongoing network activities and performance metrics, it will streamline the alarm handling process, enabling faster response times and reducing system downtime. Using the Architectural Framework introduced by project C24.0.648, this Phase II Catalyst aims to achieve the following: 1- Create a Knowledge Graph of the entities and relationships between network elements, network topology, network faults, and inventory data, and leverage it for AI operations that support predictive actions, fault correlation and assessment, and closed-loop automation. 2- Integrate TM Forum APIs within end-to-end service assurance workflows that automate manual processes in line with Autonomous Network principles. 3- Enable intent-based RAN self-healing and self-optimization using event-driven automation in conjunction with a Mixture-of-Agents pattern that employs the Knowledge Graph defined under point 1. This involves deciding on the next best action, issuing service requests, and managing element configuration as part of closed-loop invocations. 4- TMF Policy Alignment using TMF Standards and Principles as mentioned below. Our vision is to empower our telecom network operators with AI-driven intelligence that not only meets the current demand but anticipates future challenges. Leveraging machine learning algorithms, advance analytics and automation, pattern correlation and the building next best action recommendations; we can pave the way for a smarter and more resilient telecom infrastructure in spite of the environmental and consequential challenges faced.