Nokia and Databricks complete test for AI platform powering next-generation autonomous networks

Nokia and Databricks Validate Cloud-Agnostic Architecture

Nokia and Databricks announced the successful completion of a joint proof of concept demonstrating a unified data platform built to enable AI-driven autonomous networks for telecom operators 1

1

. The collaboration addresses a persistent industry challenge where telecom operators struggle with hundreds of siloed operational systems, each maintaining separate data architectures that prevent consistent AI deployment across network domains 2

2

. Engineering teams from both companies focused on validating a cloud-agnostic approach that allows network operators to deploy real-time analytics at scale without rewriting code across different environments.

Real-Time Performance Management at Tier-1 Scale

The technical work centered on a real-time performance management use case, simulating analytics ingestion at tier-1 operator scale in the cloud 1

1

. Data pipelines were created once and deployed across different platforms without modification, with the same workflows running on both Databricks and an open-source stack based on Apache Flink, Kafka, and Iceberg. This architecture supports real-time streaming, batch processing, and query-time data products, establishing the data foundations necessary for next-generation autonomous networks 3

3

. Nokia engineers developed transformation logic using platform-independent Python expressions, while the teams validated a custom compiler that automatically adapted workflows at deployment, translating abstract logic into native formats such as Delta Live Tables for Databricks or Flink SQL for open-source systems.

AI Agents Enable Automated Data Product Creation

The project demonstrated AI-powered creation of new data products using natural language prompts, where an intelligent data fabric agent can generate new data products, request validation, and deploy pipelines automatically 1

1

. According to the companies, AI agents can create and deploy new data services with limited manual intervention, enabling cross-domain decision-making across network operations 3

3

. This capability matters for operational efficiency as it reduces the time and technical expertise required to adapt network analytics in response to changing conditions.

Industry Leaders Signal Long-Term Collaboration

"Teaming up with Databricks represents a big step as we work toward building the types of data foundations required for next-generation autonomous networks," said Oguz Sunay, CTO AI and Autonomous Networks at Nokia 1

1

. Nevash Pillay, Global Head of Telecommunications Industry at Databricks, added that "our collaboration with Nokia demonstrates how a unified data platform can help simplify operations and unlock the value of AI across network domains" 1

1

. Nokia and Databricks plan to continue collaboration on enhancing autonomous network capabilities, with a focus on developing systems where AI applications can analyze and act on large volumes of data in real time 3

3

.

What This Means for Telecom Operators

The successful proof of concept signals a path forward for telecom operators seeking to modernize legacy infrastructure without wholesale replacement of existing systems. By enabling cloud-agnostic deployment, operators gain flexibility to choose infrastructure providers based on cost and performance rather than being locked into specific vendor ecosystems. The ability to deploy the same data flows across multiple platforms reduces development costs and accelerates time-to-market for new AI-powered network services. Short-term implications include faster deployment of real-time analytics for network optimization, while longer-term benefits point toward fully autonomous networks capable of self-healing and dynamic resource allocation without human intervention. Industry observers will be watching whether this AI platform for autonomous networks can scale beyond proof of concept to production deployments at major carriers, and whether the approach can extend beyond performance management to other critical network functions like security and capacity planning.