AlOps, agentic NOC, fault management, and performance intelligence, bridging OSS complexity with autonomous decision-making.
Super-apps, customer portals, BSS modernization, and Al-driven digital experiences that redefine how operators engage subscribers.
Proprietary Al accelerators (NIA, Genome Al), data engineering platforms, and reusable frameworks built for telecom scale.
From manual war rooms to an AI-native immune system - where agents sense, reason, and heal across RAN, transport, and core without human intervention.
Service and Business/customer impact assessment
MTTR reduction
Fewer incidents
Faster fault identification
Visit the official TM Forum Catalyst page
Despite a decade of automation investment, most NOCs still depend on manual incident handling. Networks generate thousands of alarms daily across RAN, transport and core — yet critical issues still take hours to resolve. With Open RAN, 5G slicing and edge computing, the manual model is no longer fit for purpose. The opportunity is to fundamentally modernise how the NOC works.
.svg)
.svg)
Define and map the SLAs, customer experience metrics and business KPIs that matter operationally — translating network events into commercial consequence.
AI agents that combine correlated faults and anomaly signals with service, customer, and topology data to estimate affected subscriber segments and severity.
Outcome-based action prioritization - which incident to address first, what mitigation is appropriate, what communication is needed.
T-Mobile USA
Axian Telecom
Turknet
Telefonica
Digital Nasional Berhad
A governed execution architecture enabling AI agents to safely discover, access, and act across real telecom systems, aligned to TM Forum ODA.
Lead architect & framework design, use cases, and ODA extension
Customer service cost reduction
Improved productivity gains
Weeks to months to deploy AI
Telecom Al is stuck between experimentation and execution because agents lack a governed, reusable way to discover capabilities, access BSS/OSS systems, validate actions, and operate with accountability. Without the right execution framework, every use case becomes a custom integration effort, slowing the path from pilot to production.
A governed execution framework that gives Al agents a safe, reusable, and auditable path to turn telecom intent into controlled BSS/OSS action.
Six governed layers - each one enforcing safety before the next can act.
The catalyst proposes a new section in ODA Canvas YAML — enabling BSS components to declare AI-callable capabilities at deployment time. A practical, standards-aligned mechanism for AI-enabling ODA without architectural disruption.
Every AI agent action is evaluated against a Cedar policy engine at the session level AND per individual tool call. Principal claims, action type, resource, and live runtime context are all validated before execution.
A complete working pattern for AI agents to discover tools, be policy-checked, interact with BSS through canonical abstractions and maintain a full audit trail; submitted to TM Forum as input to future agentic execution standards.
SK Telecom
Telefonica Germany GmbH
Deutsche Telekom AG
Globe Telecom
5G slicing, Open RAN and cloud-native cores have made networks too complex to manage manually. Al is not optional — it is the only way to scale operations.
CSPs face margin pressure with flat ARPU and rising OPEX. Al-driven NOC and governed Agentic Al are the lever that shifts operations from cost centre to competitive asset.
TM Forum's ANL4 and ODA roadmap are the industry north star. Both catalysts position Nagarro at the frontier, as a contributor, not just a consumer, of standards.
