OpSight

Projects

AI in production,
in real industry.

Not lab proof-of-concepts: systems that operators, engineers and supervisors use every day. For confidentiality we do not name our clients — we tell you the problems, the solutions and the status of each project.

In production Tissue manufacturing · Multi-plant · Europe and USA

MRO Intelligence for a global tissue manufacturing leader

The challenge

Thousands of industrial spare parts to identify and code into the catalogue: worn nameplates, operators' tacit knowledge, hours of manual ERP searches and duplicate codes across plants in Europe and the United States.

The solution

OpSight Maintain — powered by Parts.AI — with an offline-first mobile app for capture on the line, an AI recognition pipeline (OCR, exact/fuzzy SAP match, merchandise classification) and a back-office validation console. Enterprise distribution via MDM, corporate SSO with Entra ID, native export to SAP.

Status

In production across multiple plants, including sites in the United States, with apps published on enterprise stores and components processed every day.

Computer VisionOCRSAP MatchAzureEntra IDiOS / Android
Discover OpSight Maintain →
Progressive rollout Shipbuilding · Superyachts · Italy

AI-for-production programme at a leading superyacht shipyard

The challenge

A project crossing years, sheds and thousands of documents: its truth fragmented across archives, spreadsheets and the memory of a few key people. Decisions hard to reconstruct, a lead time nobody sees in full.

The solution

A multi-year programme built around a digital spine of the project — the 4D Ship Plan: every document and decision anchored to the vessel's breakdown and to the programme phases, from design to launch to delivery. Vertical AI modules plug into the spine in progressive rollout; the LLM proposes, the deterministic engine decides.

Status

Platform live with corporate SSO; modules in progressive rollout according to the programme plan.

4D Ship PlanDocument AIRAGConstraint solverAzureSpec-driven
Discover OpSight Execute →
Pilot programme Manufacturing · Operations · Finance

Inventory Discrepancy Intelligence

The challenge

Warehouse discrepancies — recurring adjustments, anomalies in SAP movements, gaps between physical and book stock — surface at year-end close, when it is too late to fix the process that generates them.

The solution

OpSight Observe: a parametric controls library over SAP MM movements, an ML normality engine that learns each plant's patterns, and an investigative copilot for root cause analysis. The journey starts with an assessment on your historical data: we show you which discrepancies could have been caught in real time.

Status

Offering in pilot programme: assessment on historical data and proof-of-value on the perimeter of one plant.

SAP MMAnomaly DetectionMachine LearningAI Copilot
Discover OpSight Observe →

The next project could be yours.

Tell us about the process you want to improve: we will honestly tell you whether and how AI can help, starting from an assessment on your real case.

Let's talk →