Deutsche Telekom: AI agents for mobile network

• AI agent "RAN Guardian Agent" now supports further improvement of network quality
• Best network experience for customers through faster analyses and solutions in network operation
• First step towards autonomous, self-healing networks

Deutsche Telekom’s AI agent, the “RAN Guardian Agent,” is now live. Operating in the context of network events and exceptional situations, the agent actively supports network management by monitoring mobile network performance, assisting in troubleshooting, and optimization. Its intelligent capabilities enable faster responses and more resilient network operations, ensuring an even better customer experience. The project was jointly presented by Deutsche Telekom and Google Cloud at the Mobile World Congress (MWC) earlier this year.

“With the introduction of the RAN Guardian Agent, we are the first network operator to rely on a highly developed AI agent in network management. Our development teams have thus done pioneering work in the development of AI agents for mobile networks,” explains Abdu Mudesir, Board member for Product and Technology at Deutsche Telekom. “With an intelligent interaction between our network experts and AI, we are solving specific challenges for the benefit of our customers - for the best network. And we are taking a big step towards autonomous, self-healing networks.”

The RAN Guardian Agent is capable of analyzing network behavior, identifying performance anomalies, and autonomously initiating corrective actions. It functions as a multi-agent system, composed of several specialized AI agents that work seamlessly together to ensure optimal network performance.
In the first step, an AI agent continuously scans and analyzes publicly available sources, such as online directories and social media, to identify upcoming public events across Germany. It compiles a comprehensive list, estimates their scale, determines their locations, and classifies them accordingly. All gathered information is then automatically verified by the agent.

In the second step, another component of the RAN Guardian uses this data to assess how well the mobile network can handle the expected traffic during each event. It evaluates the capacity of nearby antennas, monitors key network parameters in real time, and can immediately recognize high utilization levels, recommending suitable optimization measures if required.

In the third step, an additional AI agent executes these measures, for example by reallocating network resources or adjusting configurations. It also documents all actions and outcomes to continuously improve performance and enhance readiness for future events.

Telecom networks are becoming increasingly complex, making traditional automation approaches insufficient for real-time problem solving. AI agents powered by large language models (LLMs) and advanced reasoning frameworks represent the next evolution, systems capable of understanding, reasoning, acting, learning, and even validating each other’s results.

The RAN Guardian Agent embodies this new generation of intelligent automation. Working in concert with various AI applications, it identifies irregularities in the network and autonomously takes corrective actions to maintain optimal performance. By supporting network management teams, it significantly accelerates troubleshooting and contributes to a consistently high-quality mobile experience. Processes that previously took roughly an hour can now be completed within just a few minutes.

The RAN Guardian Agent was created with Google’s Gemini 2.5. In addition, it also uses CloudRun, BigQuery, and Firestore. This enables the following functions:

• Autonomous RAN Performance Monitoring: The RAN Guardian continuously analyzes key network parameters in real time to predict and detect anomalies.
• AI-driven problem classification and routing: The agent identifies and prioritizes degradations based on multiple data sources, including network monitoring data, inventory data, performance data, and coverage data.
• Proactive network optimization: The agent recommends and autonomously implements corrective actions, including resource reallocating or adjusting the network configuration.