Last year, I was on vacation in Goa in India with my brother. We ended up renting a motor scooter and buying exactly two liters of gas in the middle of nowhere (pictured above). The shop gave us two plastic water bottles filled with gasoline. Ever since, I’ve wondered how a remote village in the middle of a rural area in India manages to get oil and gas.
There are some interesting logistical challenges with major hubs and last-mile deliveries. But if you take a step back and look at a macro level, it seems quite astonishing that most countries never run out of oil and gas. Every day, the world consumes 95 million barrels of oil. Somehow, a tiny shop in a tiny village still has gas for you.
And that’s why Kayrros wants to make it easier to predict oil and gas consumption, production and storage. Think about it this way — Kayrros provides a weather forecast, but for oil and gas.
Right now, the startup is only testing the product with beta testing clients. Kayrros plans to recruit 40 new employees, and it’s going to be a tough job as I’m sure they’ll need a ton of data scientists and other AI engineers to come up with predictive models.
According to Les Echos, Kayrros aggregates a ton of data to predict oil and gas consumption, such as road traffic, customs data, pipeline and oil tanker information, satellite images and social network data. Like with most prediction models, you need as much data as possible to make accurate predictions.
The good thing is that the energy market is a very lucrative one. Oil and gas companies are some of the largest companies in the world, and many financial traders specialize exclusively on oil and gas. They also want as much data as possible to make the right decisions when it comes to logistics, infrastructure, financial gain and more. Kayrros could end up selling some expensive contracts.