Success Story: Telefónica Global Solutions

Telefónica Global Solutions drives network automation with GenAI Platform and Altostratus

On its journey toward a smarter network, Telefónica Global Solutions (TGS) partnered with Altostratus, part of Telefónica Tech and an AWS Premier Partner, to develop a Proof of Concept (PoC) exploring the potential of Generative AI.

Through GenAI Platform Altostratus’ Generative AI based solution TGS transformed how network operators query, diagnose, and manage the network, bringing greater agility, efficiency, and security to its operations.

The challenge: integrating Generative AI into a complex global network

Telefónica Global Solutions manages an international network with thousands of elements and distributed data sources. In this context, the company set out to integrate Generative Artificial Intelligence into its automation strategy to accelerate incident resolution and optimize network operations.

The project needed to enable querying of network status (latency, incidents, comparisons), diagnosing configuration deviations against technical standards, and consolidating dispersed information from AWS services (S3, OpenSearch) and a private internal Wiki, which was previously inaccessible due to private IP restrictions.

The challenge was clear: to automate network operational knowledge, reducing reliance on manual searches and improving responsiveness to incidents.

GenAI Platform: Altostratus’ solution to accelerate intelligent network automation

Altostratus designed and executed a 16-week Proof of Concept, deploying GenAI Platform within Telefónica Global Solutions’ AWS account. The solution integrated internal data sources and AWS managed services, leveraging Amazon Bedrock for generative model management and AWS Lambda for workflow and query automation.

As part of the project, two specialized chatbots were developed to enhance network operations and technical support:

  • Operational support chatbot: responds to daily queries by analyzing logs in Amazon OpenSearch and cross-referencing them with documentation.
  • Diagnostics and configuration chatbot: detects deviations between the actual network state and standard configurations, providing step-by-step instructions for remediation.
Gen AI Platform

Technical highlights

During the Proof of Concept, GenAI Platform proved its ability to integrate into complex environments. The solution successfully connected to a private Wiki via private IP or FQDN, overcoming access limitations and enabling secure, efficient use of internal knowledge.

Throughout the project, Altostratus supported Telefónica Global Solutions at every stage from KPI definition to integration and technical training ensuring a smooth adoption process. Additionally, A/B testing with multiple Generative AI models was conducted to optimize accuracy and system performance.

A successful pilot paving the way for intelligent automation

The Proof of Concept (PoC) was successfully completed and validated by Telefónica Global Solutions, meeting all defined objectives in terms of time, scope, and budget.

  • Model accuracy: achieved target precision levels for support and diagnostics tasks, ensuring consistent and reliable responses.
  • Performance and scalability: demonstrated optimal response times and the ability to process large data volumes in distributed environments.
  • Cost control: maintained cost per execution unit within the agreed framework.
  • On-time delivery: all 16 weeks of development, documentation, and training were completed as planned.

The result was a technical and functional validation demonstrating how Generative AI can be effectively integrated into critical network environments, reducing diagnostic time and increasing operational support efficiency.

This success marks the beginning of a new phase for TGS, advancing alongside Altostratus and AWS toward a large-scale intelligent network automation strategy.

Privacy Overview

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.