Jua: extremely accurate weather predictions

“What drives us emotionally, what we will put decades of our lives into researching is the idea that we can significantly improve how humans interact with the environment,” says Jua co-founder and CEO Andreas Brenner.

Andreas explains that with old-school technology, weather forecasting is limited to general geographic area information. “But customers want to know: for my field, should I add fertilizer? For my wind-farm, how much power will it produce? So my co-founder Marvin Gabler asked himself, why couldn’t we offer more precise forecasts?”

Founded in May 2022 with the aim of providing faster, more efficient and more accurate weather prediction models, Berlin-based Jua has made significant strides within a relatively brief period. In January 2023, they launched the first global weather forecasting system based completely on machine learning. With tens of millions of primary data points and an end-to-end AI-based weather model, Andreas says that their platform can predict weather with an astonishing level of accuracy and speed.

The benefits of better weather forecasting are many and significant, particularly as climate change hastens the unpredictability of weather patterns across the globe. As Andreas notes, “Half of the world’s population doesn’t have access to reliable weather forecasts. And in locations where useful forecasts exist, there is significant room for improvement.” Timelier, more precise forecasts have the potential to foster greater weather resiliency and climate adaptation as well as to contribute to the global economy through efficiency gains across a range of sectors, from agriculture to energy, or even insurance. Although it mainly serves European power and energy sectors, Jua maintains outward-looking ambitions, which are reflected in their name itself: “Jua is the Swahili word for sun. I think that, by building a global system, some of the biggest benefactors will be in developing countries.”

With an equity investment from Orbital Ventures, backed by the EIF, Andreas and his team were able to lift the company to the next level. “We had built a lab prototype, but to grow it into a commercially viable product, we needed VC funding.“ Despite the challenges of forecasting, one thing is certain for Andreas: “Everyone is dependent in some way on the weather, which is why the transition to machine learning is not a question but a necessity.”