alt Hacker NewsIf you are reviewing the changes and keep giving input, it is not that token consuming. But many people go with full yolo with agent mode.
If you are reviewing the changes and keep giving input, it is not that token consuming. But many people go with full yolo with agent mode.
I honestly felt I was going pretty YOLO :)
I write specs and review, but the LLM writes all the code, tests, etc and only gets re-prompted when it screws up badly. That said, I'm not running that many agent loops in parallel, so maybe thats where the cost bites