The traditionally highly manual process of wine production is getting a digital makeover in the form of innovative AI and IoT-driven technology.
Winely is the New Zealand-based developer of an AI-powered data intelligence platform for monitoring the development of wine fermentation. Breakthroughs in its cloud-based technology, which uses IoT data from in-tank sensors, mean that winemakers can now autonomously monitor and manage the fermentation status of their batches. The new technology is to be pioneered by a selection of pilot customers in California’s wine sector.
“We couldn’t be more excited to debut the next iteration of our product,” said Abbe Hyde, Co-founder and Chief Product Officer of Winely. “The new enhancements allow for a ‘set & forget’ experience, meaning our customers can easily self-install when transferring Winely’s device from tank to tank. Once positioned in the tank, they won’t ever have to think about the sensor again, as it can withstand even the most rigorous red wine ferment and high-pressure conditions. These create extremely difficult environments for technology to survive in, due to the intense pressures of reds, which is a problem that the industry has been trying to solve for the better part of two decades now.”
Previously, the only way to gain insight into fermentation was to put very expensive and hard-to-come-by precision sensors into a tank, generally limited to one sensor per attribute. Winely’s new technology is able to capture millions of data points holistically at a low cost, while applying machine learning to that data to enable predictive vision on each ferment, enabling winemakers to get ahead of the game on each tank.
“Our approach to this enhancement has been unique,” Hyde added. “The default approach by others has been to ‘just try a new sensor.’ Instead, our approach is to upgrade our algorithms. This means we can continue to iterate our product and improve its accuracy, even in the middle of a vintage, whilst the device is still in the tank — thereby enabling continuous upgrades through Cloud-based machine learning updates, rather than throwing away the hardware and starting discovery from scratch.”
Responses