Lore

Smoothing the Curve (Always-On AI Usage)

Overview

The practice of keeping token/usage activity consistent over time — avoiding dips on weekends, holidays, or vacations — so AI agents keep working autonomously while humans are away, rather than company-wide AI usage pausing whenever people log off.

Application

Monitor usage trend lines (see Token-Usage Leaderboard & Six-Tier Framework) for dips correlated with employee absence, and configure agents/automations to keep running during those gaps. The point is not to discourage vacation but to reframe the metric: a flat or rising curve during a lull signals AI filling in for absent humans, not humans being pressured to stay online.

Quote

"I want to make sure people did not miss it of smoothing the curve not because you don't want people to take vacation, but when people are taking vacation, you want AI to be filling in in the gaps and working autonomously." — John Kim

Related

Chunking to Avoid Cloud Anxiety

An alternative framing: rather than running agents continuously and feeling guilty during any gap where you're not "kicking off agents," deliberately chunk AI usage into discrete blocks. This is reported as both more enjoyable and more productive, and it directly counters the burnout some AI-heavy coders describe as "cloud anxiety."