Nokia and Databricks Validate Cloud‑Agnostic Data Platform for Autonomous Telecom Networks

The proof of concept validates a compiler-driven, vendor‑neutral architecture that aims to run the same AI workflows on cloud platforms or open‑source stacks without rewriting code.

Overview

Nokia and Databricks completed a joint proof of concept on June 24 that showed a unified, substrate‑agnostic data platform built to support AI-driven autonomous network operations.
Engineers used an abstract Python transformation layer plus a custom compiler to translate platform‑independent logic into native runtimes such as Delta Live Tables or Flink SQL so pipelines could be deployed unchanged across Databricks and an open‑source stack using Apache Flink, Kafka, and Iceberg.
The test demonstrated an intelligent data‑fabric agent that can create new data products from natural‑language prompts, request human validation, and automatically deploy pipelines to speed operations and reduce manual work.
Trials simulated real‑time performance‑management at tier‑1 scale and validated query‑time data products and zero‑copy sharing to enable cross‑domain analytics without duplicating large datasets.
Hours after the PoC was announced, Nokia said it expanded its collaboration with AWS to run Nokia’s Autonomous Networks Fabric on AWS with availability expected later in 2026, and market reporting recorded a modest same‑day uplift in Nokia’s stock.