Angelina YangFounder and Host, TwoSetAI · AI builder for visibility and discoverability

Your Overnight Training Jobs Are Greener Than You Think

11 Mar 2026 · 1 min

Watch on YouTube

0:37

In this brief dialogue, Angelina Yang explains that she often schedules machine learning training runs overnight to take advantage of lower costs and increased GPU availability. The other speaker expands on this by noting that nighttime electricity is frequently less carbon‑intensive because modern grids are dynamic: the energy mix shifts with demand. Using Denmark as an illustration, they describe how abundant wind power makes the night‑time grid almost entirely green, whereas daytime demand forces the grid to rely on biomass or fossil‑fuel generators. The exchange highlights that aligning ML workloads with off‑peak hours can simultaneously reduce expenses and lower the carbon footprint of training jobs.

Chapters

Sometimes I try to kick off training overnight because it's cheaper and I can fight for more GPU space.
So what you say, when it is cheaper at night, is also less carbon intensive because our electricity grids are, they're dynamic. So depending on the demand that we have, the energy mixture that goes into producing electricity changes. Say for instance, in Denmark, on a windy day, we have a lot of wind energy here. So then late in the night it's entirely green, but then during the daytime when there's more demand, just the wind energy cannot, you know, fulfill the needs. So then there's a lot of biomass or other fossil fuel-based energy that has to come into the grid.