Antioch's Series A is a bet that robots learn faster in software. Software teams ship a change, run the tests, and know within minutes whether it worked. Physical AI teams have spent years waiting on hardware, because every new perception model meant another fleet run, another sensor rig, another week of collecting data on a real road or a real farm. That gap is the whole company. Antioch's pitch is that the fleet moves into simulation. Spin up many virtual copies of the machine with…