Understand this story
On November 12, 2026, Bright Machines announced a real-time data dashboard solution for air-gapped factories: at the factory edge, Debezium captures PostgreSQL changes and writes them to Apache Iceberg, which are then transmitted out through a single auditable outbound path and read in place by Databricks via Unity Catalog federation and AWS Glue, giving customers real-time dashboards without needing to open inbound access to the factory. Data written on the production line can be queried in Databricks within 1–2 minutes, dashboards refresh every 10 minutes, and heartbeat checks can detect pipeline failures within 30 minutes (previously this took days); the first customer dashboard went live about two months after project kickoff. The solution replaces the previous nine loosely coupled services with a single governed platform.
AI-generated from public reports
Developments
1 confirmed developments
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 12
Databricks BlogFirst-party
Air-Gapped but Not a Data Black Box: How Bright Machines Gives Customers Real-Time Dashboards Without Opening the Factory
At the factory edge, Bright Machines uses Debezium to capture PostgreSQL changes and write them as Apache Iceberg, which are then transmitted through a single auditable outbound path and read in place by Databricks via Unity Catalog federation and AWS Glue, giving customers real-time dashboards without requiring inbound access to the factory. Data written on the production line becomes queryable in Databricks within 1–2 minutes, dashboards refresh every 10 minutes, and heartbeat checks can detect pipeline failures within 30 minutes (previously taking several days); the first customer dashboard went live about two months after project kickoff. The solution replaced nine previously loosely coupled services with a single governed platform.