Vodafone and Zinkworks are collaborating on a generative AI platform designed to simplify telecom management and reduce the operational challenges of 5G networks. The platform aims to improve network performance and reduce energy consumption by making it easier to build and deploy Radio Access Network (RAN) applications.
Managing modern 5G networks requires highly specialised knowledge that is often split across different teams. Traditionally, creating and operating RAN applications has required deep radio engineering expertise and advanced software development skills. This separation can slow innovation, extend delivery timelines, and increase costs.
Vodafone and Zinkworks’ Rapid RIC platform tackles that complexity by enabling radio engineers to create operational apps without needing to be expert coders. The platform provides a visual, drag-and-drop interface combined with generative AI that automatically generates, validates, and assembles the underlying code for those apps.
Vodafone’s AI adoption and what it means for modern telcos
Rapid RIC blends Vodafone’s data capabilities with a user-friendly visual programming environment and AI-driven code generation. Engineers can design application workflows by dragging and dropping components; the platform then writes and checks the code, streamlining the entire development cycle.
By removing the need for manual model training and extensive data preparation, Vodafone expects development cycles to shrink dramatically. The company projects it can create and deploy new RAN features in three to four weeks rather than months, cutting development time for RAN apps by an estimated 60–70 percent.
Shorter development times translate into practical benefits for both the operator and its customers. Rapidly deployable rApps can automate tasks that are currently manual: boosting signal strength where needed, diagnosing and fixing issues remotely, and switching off unused connections to conserve energy.
End users benefit from faster internet speeds, improved coverage in congested urban areas and underserved rural locations, and fewer service interruptions.
The project is expected to be operational by early 2026, and Vodafone is already developing two initial rApps at its research centre in Málaga, Spain.
The first app focuses on energy savings by automatically powering down idle mobile connections and ensuring those connections reliably restart when required—addressing a common source of poor user experience. The second app uses machine learning to refine radio coverage parameters that often remain at default factory settings, improving overall network efficiency and performance.
Alberto Ripepi, Chief Network Officer at Vodafone, said the platform is intended to “simplify and accelerate the deployment of AI-powered applications that directly improve the customer experience,” enabling stronger, more reliable signals, greater capacity, and progress toward sustainability goals.
How it works and the challenges
Rapid RIC is designed for modern software-driven 5G networks and is particularly suited to Open RAN deployments, which encourage innovation by allowing operators to mix and match equipment and software from different vendors at a single base station.
A critical challenge for deploying AI-generated code in national telecommunication networks is safety and reliability. Introducing new software that can control network behaviour must be carefully validated. To address this, Rapid RIC uses an AI-powered simulator to test rApps before they are deployed and to continuously monitor their performance in production.
This simulation-first strategy offers a practical balance: it provides realistic testing and feedback on new configurations without relying solely on outdated historical data or requiring the expense and complexity of maintaining a full digital twin of the entire network.
Technically, the platform will leverage Vodafone’s existing data infrastructure and will run primarily on the operator’s secure Google Cloud Platform (GCP) environment. This private cloud setup lets Vodafone teams upload, monitor, and deploy rApps across multiple European markets from a centralized, secure location.
Paul Madden, CEO of Zinkworks, commented that the collaboration represents a significant advance in how Open RAN applications are developed and deployed, and praised the visual programming tools that support faster innovation, quicker service upgrades, and improved network responsiveness.
Vodafone plans to retain ownership of the platform while offering related apps and services—such as software testing and monitoring—to other industry participants, extending the commercial reach of its network expertise.
Automating 5G network operations at scale
The Vodafone–Zinkworks partnership illustrates how generative AI is moving from content creation into operational automation at scale within telecoms. The approach demonstrates practical ways AI can accelerate network improvements and drive measurable business outcomes.
Key takeaways for telecom leaders include:
- Use generative AI to bridge skill gaps: The platform empowers radio engineers to solve operational problems directly without relying on separate software teams. Organisations should identify similar cross-discipline bottlenecks where generative AI can enable domain experts to act independently.
- Prioritise practical testing: Deploying AI-generated code in live networks carries risk. Vodafone’s use of an AI-driven simulator shows how operators can validate changes safely and cost-effectively without maintaining a full-scale digital twin.
- Align AI with business goals: Rapid RIC ties AI capabilities to clear business outcomes: faster development cycles, improved customer experience, and reduced energy consumption. AI initiatives should be framed around measurable business value to secure executive support.
- Build on existing infrastructure: The platform leverages Vodafone’s existing GCP-based data environment, showing that effective AI deployments often extend and integrate with current secure cloud platforms rather than requiring a complete rebuild.
Vodafone is effectively building an “AI factory” to convert engineering knowledge into automated software that scales across its network. This marks a shift from experimental AI tools toward operationalising AI in critical parts of the business to deliver real customer and sustainability benefits.
See also: AI regulations in telecoms: Navigating the complex web
Want to learn more about AI and big data from industry leaders? Attend the AI & Big Data Expo events held in Amsterdam, California, and London, where industry experts present practical insights on AI, data and security.
Telecoms is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars through industry event listings.