Tech Mahindra has announced a partnership with NVIDIA to introduce an AI-powered Telco Network Operations Reasoning Agent designed to help Communication Service Providers accelerate their transition toward Level 4+ autonomous networks. The new solution — announced at the Mobile World Congress (MWC 2026) — aims to transform traditional Network Operations Centers into intelligent, closed-loop operational environments powered by advanced artificial intelligence.
The AI-driven solution is delivered through Tech Mahindra’s Orion platform, integrating AI engineering, telecom domain expertise, and a human-in-the-loop design approach. Built with enterprise-grade trust, the platform ensures that customer and Personally Identifiable Information is not used, enabling CSPs to deploy AI while maintaining strict data privacy and governance standards.
AI Reasoning to Transform Network Operations
Network Operations Centers have traditionally relied on rule-based and open-loop workflows that require extensive manual intervention. Engineers often spend significant time correlating alarms, logs, and performance data across multiple systems, which slows incident resolution and affects operational efficiency.
The collaboration embeds AI-driven reasoning directly into core network operations, enabling closed-loop execution. The AI reasoning agent supports autonomous alarm validation, root cause analysis, and automated resolution across Operations Support Systems and Business Support Systems environments. This results in faster incident response, improved operational consistency, and enhanced customer experience.
A key differentiator of the solution is its extensibility. Once CSPs deploy a foundational Large Telco Model, they can scale domain-specific reasoning agents tailored to their operational environments. The system integrates seamlessly with existing data lakes, tools, and workflows, enabling incremental AI adoption without disrupting legacy systems.
Powered by NVIDIA AI Enterprise
Leveraging NVIDIA AI Enterprise software, Tech Mahindra customized a reasoning model using synthetic and anonymized data through NVIDIA NeMo and deployed it as an NVIDIA NIM inference microservice. The fine-tuned model demonstrated a 2 to 3 times improvement in accuracy compared to non-fine-tuned models, strengthening real-time decision-making capabilities within live NOC environments.
By combining NVIDIA’s AI software stack with Tech Mahindra’s telecom expertise, the solution enables reasoning-based AI systems that can act, adapt, and learn dynamically in complex network scenarios.
Advancing Toward L4+ Autonomous Networks The move toward autonomous networks is an iterative and ecosystem-driven process. Modular architecture and reusable AI pipelines allow CSPs to adopt reasoning capabilities gradually and scale them across their operations. The collaboration underscores Tech Mahindra’s focus on advancing AI innovation in telecom through intelligent automation, deep learning, and multimodal AI models.
SHAFANA FAZAL
