When AI Hits Its Daily Token Limit, What Should Engineers Do?
Is the industry ready for tokens-constrained work?

A consultant's story about running out of AI tokens mid-day sparks a debate: what are engineers supposed to do when their AI agents can't work? The author argues that picking up an agent's work manually is impractical, and that filling the remaining hours with 'useful' tasks like studying or shadowing colleagues is noble but not universally accepted. With companies increasingly coupling work to time, the real question is whether we're ready for a world where token budgets, not time, constrain productivity—and whether more tokens just encourage sloppier work.
If in the '80s we fought to squeeze a program into a few kilobytes, we now have to squeeze the most useful LLM work into the tokens we can afford.