Weather risk is hyperlocal.
AI weather models are getting cheaper, faster, and more available. That is not the bottleneck anymore. The bottleneck is knowing what will happen over a specific asset in the next minutes.
Energy traders, airports, insurers, telecom operators, and satellite companies do not lose money because the atmosphere is unknown in general. They lose money because a specific cloud appears over a specific asset at a specific time.
- A cloud shadow becomes lost solar revenue.
- A low ceiling becomes a grounded aircraft.
- A convective cell becomes hail damage.
- A blocked sky becomes a failed optical link.
That layer is still missing.
Matics Labs builds the physical AI sensing network for asset-level weather nowcasting. We start with clouds, the highest-value source of short-term atmospheric risk. The company that owns local atmospheric sensing owns the correction layer above global models.
Ground first, air next, space later.
The goal is the atmospheric intelligence layer for asset-level decisions.
Starting with solar intraday trading.
Solar volatility is not a forecasting problem. It's a control problem.
A cloud field appears upstream. PV drops within minutes. Intraday prices move. A balanced position becomes an exposure.
Matics Labs turns local sky intelligence into action.
We combine cloud sensing, solar nowcasting, price and imbalance forecasting, battery state, and execution guardrails into one operating loop.
For utility-scale PV, co-located batteries, BRPs, VPPs, renewable traders, and large C&I sites with solar PPAs.
Fewer open deviations. Better battery dispatch. Higher intraday capture. Lower imbalance cost.
Every few minutes, the system asks: what is the best risk-adjusted action before the market moves?