Mavenir launches AI framework to automate network fault resolution

Mavenir has introduced a multi-agent AI system designed to automate network operations by converting operational knowledge into repeatable workflows, the company announced ahead of DTW Ignite 2026 in Copenhagen.

The Agentic Service Assurance Framework addresses a pressing challenge for telecom operators: increasingly complex networks coupled with a shrinking pool of experienced engineers. The system captures expertise from live Network Operations Centre teams and third-party vendor operations, then converts that knowledge into a vendor-neutral automation catalogue.

At the core sits an intent orchestrator that translates operators' desired outcomes—expressed in natural language or voice—into automated actions. The framework handles detection, diagnosis, recommendation, and resolution of network faults across 5G, cloud-native, and IP domains.

How the system works

The framework operates across three operational layers, spanning basic monitoring through full remediation. It integrates with existing OSS AI agents using the Agent-to-Agent protocol, allowing operators to adopt the technology without replacing current infrastructure.

Every automated workflow undergoes validation before autonomous execution, maintaining consistency and creating a full audit trail. The system aligns with TM Forum standards IG1251 and IG1453, enabling deployment across multi-vendor and multi-domain environments.

"Every operator I talk to is wrestling with the same paradox: their networks are becoming more complex while the operational expertise to run them is retiring," said Bejoy Pankajakshan, Chief Technology & Strategy Officer at Mavenir. "Intent Ops solves that paradox directly. It captures what your best engineers know, validates it, and makes it available to the entire operations organization, autonomously and at scale."

Why it matters

Network operators face mounting operational costs from fragmented systems that struggle to diagnose root causes at scale. Traditional assurance tools often flag problems without resolving them, while the engineers who understand complex network behaviors are retiring. Mavenir's framework represents an attempt to institutionalize that departing expertise before it's lost, potentially reducing the time operations teams spend chasing faults while improving overall network reliability. The standards-based approach also addresses a key industry concern: avoiding vendor lock-in while moving toward autonomous operations.

Andy Tiller, EVP Member Products at TM Forum, noted that operators need practical routes to autonomous networks that function across complex, multi-vendor environments. He characterized Mavenir's approach as demonstrating the industry direction: open, standards-based autonomy that operators can trust.

The framework's knowledge capture capability continuously learns from live NOC workflows, building an expanding automation catalogue that works across geographically distributed, cloud-native networks.

Details were first reported by Telecoms.com.