DTW-Ignite 2027
Autonomous Network Operations
Agentic NOC: Broadening the ecosystem - Phase II
URN C27.0.035
Every incident makes the network smarter: a self-learning, cost-aware agentic NOC from cell site to core.
Project companies
Agentic NOC: Broadening the Ecosystem is a Catalyst project focused on advancing autonomous network operations through AI-powered agents that can detect, diagnose, predict, and resolve network issues across the entire telecom network, from the radio access network (RAN) and transport layers through to the core network.
Building on the award-winning first phase, the project aims to create a self-learning Network Operations Center (NOC) that continuously improves through every incident it handles. The project addresses one of the telecommunications industry's biggest challenges: managing increasingly complex networks while maintaining service quality, controlling operational costs, and meeting customer expectations.
Today, many AI-driven operational solutions work only within isolated network domains, limiting their effectiveness. This Catalyst extends AI autonomy across multiple domains, enabling operators to identify root causes faster, understand service and business impacts, and proactively prevent issues before customers are affected.
At the heart of the solution is a dynamic knowledge graph and digital twin that combine network topology, operational data, historical incidents, and service information into a single intelligence layer. AI agents use this foundation to collaborate, learn from past incidents, simulate remediation actions before implementation, and make more informed decisions.
The project also introduces agent observability and tokenomics capabilities, helping operators understand both the effectiveness and cost of AI-driven operations. A unique aspect of this phase is the exploration of NOC-to-NOC collaboration, where AI agents operating in different service provider environments can securely share context and coordinate responses to incidents that span multiple networks.
The initiative is aligned with TM Forum's Autonomous Networks mission and contributes to the evolution of Open Digital Architecture (ODA), Open APIs, and next-generation autonomous operations frameworks.
Building on the award-winning first phase, the project aims to create a self-learning Network Operations Center (NOC) that continuously improves through every incident it handles. The project addresses one of the telecommunications industry's biggest challenges: managing increasingly complex networks while maintaining service quality, controlling operational costs, and meeting customer expectations.
Today, many AI-driven operational solutions work only within isolated network domains, limiting their effectiveness. This Catalyst extends AI autonomy across multiple domains, enabling operators to identify root causes faster, understand service and business impacts, and proactively prevent issues before customers are affected.
At the heart of the solution is a dynamic knowledge graph and digital twin that combine network topology, operational data, historical incidents, and service information into a single intelligence layer. AI agents use this foundation to collaborate, learn from past incidents, simulate remediation actions before implementation, and make more informed decisions.
The project also introduces agent observability and tokenomics capabilities, helping operators understand both the effectiveness and cost of AI-driven operations. A unique aspect of this phase is the exploration of NOC-to-NOC collaboration, where AI agents operating in different service provider environments can securely share context and coordinate responses to incidents that span multiple networks.
The initiative is aligned with TM Forum's Autonomous Networks mission and contributes to the evolution of Open Digital Architecture (ODA), Open APIs, and next-generation autonomous operations frameworks.
The team
Join this project
Contact us to find out how