Klyff provides a unified platform for smart manufacturing that connects heterogeneous shop‑floor devices—sensors, PLCs, vision systems, and edge hardware—and runs AI models at the edge to deliver a live, single source of truth across multiple sites.
Funding
Funding not disclosed
Founders
Product
Problem
Manufacturers struggle with fragmented data from heterogeneous shop‑floor devices—sensors, PLCs, vision systems, and edge hardware—requiring multiple dashboards and preventing a real‑time, single source of truth. This fragmentation leads to missed defects, unplanned downtime, and difficulty standardizing operations across multiple sites while keeping sensitive production data on‑premises.
Solution
Klyff offers a unified smart manufacturing platform that connects diverse edge devices using common industrial protocols (MQTT, CoAP, HTTP) and streams clean telemetry to a central hub. The platform runs AI models at the edge for quality inspection, predictive maintenance, and federated learning, delivering live metrics such as throughput, OEE, and asset health to plant managers. By keeping raw data on‑site, Klyff enables defect detection, downtime reduction, and process optimization without centralizing sensitive information. Model improvements can be shared across plants, allowing rapid rollout of new AI capabilities while maintaining data residency compliance. The result is measurable ROI in weeks, with higher first‑pass yield, reduced scrap, and increased equipment uptime.
Target Audience
Primary customers are plant managers and operations teams at mid‑size to large manufacturers in sectors such as electronics, automotive, and pharmaceuticals, as well as corporate engineering groups overseeing multiple production sites.
Features
- Multi‑protocol device connectivity (MQTT, CoAP, HTTP) for sensors, PLCs, vision systems, and edge hardware
- Edge‑deployed AI models for automated defect detection, predictive maintenance, and federated learning
- Real‑time dashboards showing throughput, OEE, FPY, alarms, and asset health on a single screen
- On‑premises data handling that preserves trade‑secret and GDPR compliance while enabling cross‑site model sharing
- Plug‑and‑play integration with existing cameras and industrial sensors, avoiding costly hardware replacements
- Quantifiable performance gains: 2–4 pp FPY improvement, 30–50 % downtime reduction, 20–30 % MTBF increase