Suggestions
Ben Toms
VP of AI | PhD | Forecasts for energy and weather derivatives
Ben Toms is the Vice President of AI at Climavision, a position he has held since April 2024. He brings extensive expertise in atmospheric science and artificial intelligence to his role. Prior to joining Climavision, Toms founded Intersphere, a company that developed an AI-based weather forecasting system, which was later acquired by Climavision.12
Educational Background
- PhD in Atmospheric Science: Colorado State University (2016-2020)
- Bachelor of Science: Dual degrees in Meteorology and Civil Engineering, with a minor in Mathematics from the University of Oklahoma.14
Professional Experience
- Vice President, AI for Weather: Climavision (April 2024 - Present)
- Founder: Intersphere (2020 - 2024)
- PhD Student & Fellow: Department of Energy Computational Science Graduate Fellow at Colorado State University.135
Toms is focused on integrating AI with traditional physics-based modeling to enhance weather forecasting capabilities, particularly for sectors such as energy and insurance. His work aims to improve the accuracy and extend the range of weather predictions.26
Highlights
A lot of people think AI isn't explainable. It is. If you build your model right, you can explain every forecast issued by AI -- for every location, every time horizon, every variable. We use these explainability methods at Climavision to identify the patterns our models use to make their forecasts. We then use these explanations to help inform how we build our models and help customers trust our forecasts. The plot below shows regions that are being used to make forecasts for the location shown by the blue dot. In this particular example, we are analyzing surface pressure patterns that are used to make a temperature forecast. The brighter regions are more important for making an accurate prediction on the date shown in the plot title. Shoot me a DM if you're interested in learning more about our explainable AI forecasts. Our AI forecasts are specifically built for longer range forecasts from three weeks out to a year.
Weather forecasts can be uncertain. Intersphere knows that, and you know that.
And this is why Intersphere's long-range weather forecasts are fully probabilistic.
What does this mean? Our forecasts say the chances that any outcome will happen.
You can use these probabilities to understand: ✔our forecasts' confidence ✔the probabilities of any event or threshold
We also provide single-value predictions based on these probabilities, such as the 50th percentile value.
And we know ease-of-access is important, so our API makes it simple to query any percentile for any location.
Interested? Contact me here or at [email protected].
The below animation shows the median of all locations' forecasts (left) and the distribution for a single location (right). The location is Houston, Texas, shown by the red dot on the map.
The animation shows a seasonal temperature forecast. All of our forecast lead times and variables have probabilistic info.
