Agent crews operate in a staged, permissioned framework that starts with analysis and recommendations and can progressively gain authority to directly control settings as trust and reliability grow.

The CEO emphasizes merging advanced AI with Sight Machine’s semantic layer to bridge general AI with challenging production data.

Autonomous AI agent crews work with Sight Machine's semantic layer to analyze raw industrial data and optimize production, targeting improvements in throughput, quality, and cost.

Sight Machine unveils autonomous agent crews for manufacturing that continuously monitor and optimize production performance, with a demonstration at Hannover Messe.

The platform is open, modular, and enterprise-grade, designed to integrate with existing architectures while preserving customer data control.

Expanded availability of the platform is planned later in 2026 as part of ongoing development and a Hannover Messe preview.

Users can access insights and recommendations through familiar tools like Excel and Teams and data platforms such as Microsoft Fabric and Databricks, reducing the need for specialized manufacturing software.

Agents use a broad analytics toolkit—statistical methods, machine learning (e.g., random forests), AI-enhanced centerlining, golden runs, cookbooks, and root-cause analytics—and communicate via the Model Context Protocol within Microsoft Azure’s secure, permissioned framework.

The Hannover Messe demonstration features Sight Machine and Microsoft’s Integrated Industrial AI Stack on Azure, showcasing an end-to-end cloud-to-edge industrial AI solution.

Operations teams can deploy and extend the semantic layer to new machines, lines, and SKUs, reducing reliance on software developers and giving teams more control.