Core Insights - A tech firm, Amperon Holdings Inc., is enhancing electric-grid forecasting by providing hourly projections of US power demand up to seven months in advance, offering a new perspective compared to traditional 15-day weather forecasts [1][2] Company Overview - Amperon, co-founded in 2018 by Sean Kelly, utilizes artificial intelligence and machine learning to generate hourly demand forecasts that are updated daily based on global weather models from Europe's largest forecasting center [2][5] - The company is backed by notable investors including Energize Capital, HSBC Holdings Plc, National Grid Plc, and Tokyo Gas [4] Industry Context - The challenge of predicting electricity usage has increased due to extreme weather events and the growing influence of heat pumps, solar energy, batteries, and electric vehicles on demand [3] - Demand from data centers is projected to more than double by 2035, increasing its share of total US electricity usage from 3.5% to 8.6% [3] Market Demand - Utilities and power retailers are seeking improved visibility into weather conditions and consumer behavior to mitigate price shocks, while energy speculators are looking for a competitive edge through advanced forecasting [4] - Amperon's forecasts are being utilized by major power companies such as PG&E Corp., Orsted AS, AES Corp., and Eversource Energy [4] Forecasting Methodology - The accuracy of Amperon's projections relies on the underlying weather forecasts, which are derived from models developed by the European Centre for Medium-Range Weather Forecasting [5][6] - Amperon's machine learning models enhance the granularity of these forecasts, providing hourly temperature predictions over a seven-month period [7] Performance Indicators - Early indications suggest that Amperon's forecasting approach can yield accurate predictions, with successful backtests showing the ability to forecast demand spikes and weather impacts well in advance [8]