Sharing idle AI rigs with friends to run agent tasks in parallel
Several people each have a machine that can run AI models, but those machines sit idle about 90% of the time. The proposed setup would check whether friends' machines are idle when a job starts, then send tasks to any available machines through something like a VPN. A setup would spread jobs across the machines instead of piling them onto one.
Agent work can often be split into parallel tasks, so this could raise total speed if the jobs are easy to divide. The main idea is to use spare local computing power before buying more hardware or paying for cloud capacity.
Key points
- The setup starts from three AI rigs that are idle most of the time.
- A job launcher would check whether friends' machines are free before sending work to them.
- routing would distribute tasks across available machines in turn.
- Parallel could run faster if they can be split cleanly.
- The cost benefit comes from using spare local compute, not from lowering token use directly.