Sight Machine Launches Agentic Manufacturing Platform That Understands Your Plant - and Improves It, Every Run

Agents work directly with your process experts to connect data, build the Semantic Model, and improve performance – deploying in days, compounding in value with every run

SAN FRANCISCO, June 11, 2026 /PRNewswire/ -- Sight Machine Inc., the leading industrial AI platform, today launched a fundamentally new version of its platform, changing how manufacturers work with AI and dramatically accelerating its impact. What typically takes months of effort by data engineers, system integrators, and dedicated specialists can now be done in days by process experts – the people who know the plant best – working directly with AI agents.

General-purpose AI platforms can't make sense of raw industrial data. Sight Machine's Semantic Model changes that: it maps every signal to an asset, every asset to a process, and every process to the KPIs an operations team actually runs the plant on – giving agents the context to investigate, reason, and act with precision.

The platform continuously finds more output, eliminates more waste, and improves with every run. Agents build tools, the Semantic Model grows, and every cycle makes the next one sharper.

Sight Machine has spent more than a decade solving the hardest problem in industrial AI: turning manufacturing data into intelligence that agents can act on. That knowledge is encoded into a single, continuously updated Semantic Model – built by process experts working directly with agents, not bottlenecked by technical intermediaries in rounds of clarification that take months and make rapid iteration impossible. Improvement compounds across the enterprise rather than being rebuilt site by site, use case by use case.

Semantic Model: Your Plant Data, Structured for Agents to Reason On

Agents connect to all OT and IT systems – sensor readings, historians, MES, ERP, documents, SOPs, and enterprise data platforms like Microsoft Fabric. They extract, map, and structure that data into one consistently organized digital model of the facility.

Blueprint, Sight Machine's proprietary small language model, automates the most labor-intensive step in any industrial data project: mapping thousands of tags to their physical assets by looking beyond field names to introspect data, recognize patterns, and infer meaning.

Engineers who know the plant define events, validate mappings, and add the tacit knowledge agents can't see. Operations teams trust the findings because they helped build the foundation the agent reasons from.

Dynamic Production: Production That Responds Continuously to Realtime Feedback

While most "manufacturing AI" sits on a dashboard waiting for someone to log in, Sight Machine agents work continuously – pursuing leads on quality drift, downtime patterns, and yield gaps, forming hypotheses, and bringing validated findings to process experts for review.

Agents investigate and propose; process experts validate and approve. Findings don't land in a report to check later – they arrive as plain-language recommendations in the tools operations teams already have open: Microsoft Teams, Microsoft Excel, a kiosk on the floor. The recommendation is specific and immediate: exactly what to do, at the moment it matters.

As agents investigate, they write reusable applications and register them for future use – so your library of capabilities grows automatically.

"Manufacturing has always had the data. What it lacked was a platform that could make that data legible to AI," said Jon Sobel, CEO and Co-Founder of Sight Machine. "Sight Machine built that platform – and built it so that the people who know the plant are the ones who run it, from day one. That's a completely different paradigm, and it's what makes improvement compound with every run."

Enterprise Agents: Plant Floor AI Unlocks Firmwide Optimization

Manufacturers are moving from data silos to data estates, yet plant floor intelligence has historically been the missing piece, trapped at the source and invisible to the enterprise systems that depend on it. Sight Machine changes that: manufacturing intelligence flows up to the enterprise, and enterprise context flows back to the plant.

Sight Machine publishes its manufacturing intelligence as an MCP server – the open standard for agent-to-agent communication – so any permissioned agent in the enterprise can access it. Push findings into Microsoft Teams, Microsoft Fabric, and Microsoft Excel. Run Omniverse simulations. Connect Databricks ML models. Intelligence flows both ways, and the compound effects reach across the firm.

Security is structural throughout: each deployment is single-tenant by design, with its own database, namespace, and keys. Manufacturers retain full control over their data.

Progressive Autonomy, at a Pace Manufacturers Control

Agents begin in investigation and recommendation mode, surfacing findings to process engineers who remain in control of production decisions. As agents demonstrate reliability, manufacturers can expand agent scope – moving progressively toward autonomous operations. The architecture is fully permissioned: manufacturers control which agents operate, what tools they can use, and how far their authority extends.

"Everything that previously required a full team – connecting data, modeling the plant, building applications – the agent now does directly with the process expert," said Andrew Home, VP of Product at Sight Machine. "What used to require a development sprint now happens in the course of an investigation. The platform scales to new use cases and lines without being constrained by team size."

About Sight Machine

The next AI revolution is in the physical world. Sight Machine is the platform built for it – connecting to all OT and IT systems, building a continuously updated Semantic Model of how the facility actually runs, and deploying AI agents that investigate production, surface improvement opportunities, and put recommendations directly into the hands of operations teams. It deploys in days, improves with every run, and integrates natively with Microsoft Azure, Microsoft Fabric, Microsoft Teams, NVIDIA Omniverse, and Databricks. Sight Machine works with manufacturers across more than 20 industries and 20 countries. Learn more at sightmachine.com.

Sight Machine Press Contact: [email protected]

SOURCE Sight Machine Inc.