How it works

From raw data to a daily recommendation in four steps.

Data collection

We pull data from multiple sources daily — historical gas prices, crude oil futures, seasonal trends, regional pricing, and experimental signals.

Each data stream feeds into its own ML model. Every model answers one question: should you buy gas today?

A second-layer model combines all predictions and learns which signals to trust based on their historical accuracy.

An LLM reviews the prediction and generates a plain-English explanation of the key factors.