Groundup.ai, an industrial machine intelligence company, launched GINA V2, the latest generation of its Cognitive Maintenance engine, at Industrial Transformation Asia-Pacific (ITAP) 2026 in Singapore on October 22, 2026. The system is designed to move industrial AI beyond simple anomaly alerts by reasoning across sound, vibration, and thermal data simultaneously. It aims to pinpoint mechanical root causes and tell operators exactly what to do before a fault reaches the plant floor.
Closing the Gap in Predictive Maintenance
Legacy predictive maintenance systems often rely on static thresholds and stop at simply flagging raw anomalies. GINA V2 combines multimodal AI, signal processing, and agentic reasoning to diagnose the specific mechanical root cause of a developing problem. The platform then prescribes precise corrective steps, placing decision-grade guidance directly in the hands of operators before an issue escalates on the plant floor.
An AI-Native Foundation
Groundup.ai has built its platform as an AI-native system from the ground up, with data, models, and agentic reasoning designed to work as one. This approach differs from much of the market, where AI is layered onto condition-monitoring architectures conceived long before modern AI existed. A clean, AI-first stack is what lets GINA move past alerting to genuine diagnosis and prescription, and it allows the platform to keep compounding as its models grow.
New Capabilities in GINA V2
The latest release identifies which machines need immediate attention by drawing on real-time sensor data, ISO guidelines, and Groundup.ai's expert AI. Each asset carries a precise health score and a traffic-light status, while a new drivetrain visualization maps sensor placement and system-level views across multiple locations. Teams can also stack sensor streams side by side, such as temperature against vibration, and model calibration through feedback helps the system sharpen over time.
Built on a Growing Asset Library
GINA V2 is supported by the Groundup.ai Asset Library, which the company describes as the world's largest industrial anomaly library. That foundation helps the system understand a customer's machines from day one instead of starting from a blank slate. Every piece of team feedback calibrates the model, so the platform becomes more accurate and useful over time.
Executive Commentary
Chief Executive Officer and Founder Leon Lim noted that maintenance software has historically been effective at signaling when something is wrong. He explained that operators have still had to determine what was failing, why, and what to do, often under pressure on machines that cannot wait. Lim stated that GINA V2 is built to answer those questions and empower the people keeping factories, fleets, and operations running to act with confidence.
Commercial Momentum and Deployment
The launch follows Groundup.ai's strongest commercial year to date, including multi-year contracts valued in the eight-figure range. The company expanded across Asia-Pacific, the Middle East, and Europe, won an open tender with PUB, Singapore's National Water Agency, and was recognized in the NextGen Tech 30 Class of 2026. GINA V2 is already being deployed across manufacturing, maritime, defense, and critical infrastructure operations.
Groundup.ai selected ITAP 2026 as the venue for the announcement because the event gathers manufacturers, utilities, and infrastructure operators across Asia-Pacific. These organizations manage aging, hard-working machinery every day, and the company sees them as the people Cognitive Maintenance was built for. Attendees can meet the Groundup.ai team at Booth A7 in the Singapore Pavilion or book a demonstration online.