Hacker News new | past | comments | ask | show | jobs | submit
veryfancy | on: LLMs can be exhausting
In agent-mode mode, IMO, the sweet spot is 2-3 concurrent tasks/sessions. You don’t want to sit waiting for it, but you don’t want to context-switch across more than a couple contexts yourself.
SchemaLoad | parent
That sounds exhausting having to non stop prompt and review without a second to stop and think.
colecut | root | parent
There is nothing dictating how long you stop and think for.
bombdailer | root | parent
Until that becomes the metric measured in performance reviews.