Blog

Updates and insights

Technical posts from the Gramm team on grid demand forecasting, deep learning architectures, and what we learn operating across U.S. power markets.

TechnicalMarch 10, 2026

Benchmarking across seven U.S. grids: what we learned

Each grid has different characteristics — solar-heavy CAISO, isolated ERCOT, wind-variable SPP. One model architecture, trained per-region. Here is what we found expanding from California to all seven.

ResearchFebruary 18, 2026

The architecture search that found something better

We benchmarked every modern deep learning architecture against seven U.S. grids. The winner was not what we expected. And by the time we submitted the paper, we had already found something more performant.

TechnicalFebruary 3, 2026

Why grid demand forecasts fail during extreme weather

Conventional forecasting methods break down during heat waves and cold snaps. The load-temperature relationship becomes nonlinear at extremes, and ISO baselines are calibrated for normal conditions.

FoundingJanuary 15, 2026

From research to product: how two arXiv papers became a forecast API

We published two papers showing modern ML architectures could beat ISO baselines by 25-71%. No one was offering this as a production API. That gap became the reason to start Gramm.

ResearchJanuary 8, 2026

Why we invented a new accuracy metric before building the model

Two models with identical MAPE can have wildly different operational consequences. We borrowed from statistical process control to define metrics that distinguish nuisance errors from dangerous ones.

FoundingDecember 12, 2025

From manufacturing floors to power grids

Gramm started in semiconductor manufacturing. Anomaly detection, root cause analysis, process deviation scoring. Then we read a CEC grant brief and recognized the problem instantly.