I Paid $60 for Two Humanoids to Clean My Home in San Francisco
A San Francisco writer hired two Tau Robotics humanoids for an hour of house cleaning and found they did most tasks well, but the booking still needed remote human operators, staff support, and a battery swap to finish the job. He argues that renting robots by the task or hour could make home robots practical, and that a shared fleet would improve them through more real-world data.
- The robots handled vacuuming, mopping, and wiping nearly as well as a human, but struggled with delicate tasks like picking up toys, socks, and pillows, and skipped the dishwasher and laundry.
- Each robot was teleoperated by a remote human, and staff handled drop-off, pickup, and shoe changes, so the real labor cost goes well beyond the $30 per robot-hour price.
- Renting by the hour lowers the commitment compared with $449 per month or $7,999 up front, and sending robots into many homes could speed up learning and improve performance.
On a Sunday, the author paid $60 for two Tau Robotics humanoids, Sergey and Chelsea, to clean his San Francisco home for an hour. They vacuumed, mopped, wiped surfaces and cleaned a toilet, covering the dining room, living room, kitchen and a half bathroom. But the robots were teleoperated: each had a human operator remotely controlling it from Tau's Mission office, and for most of the hour no staff member was in the house. The job also left some things undone. The dishwasher was not unloaded, laundry was not folded, and the robots needed a battery swap after roughly an hour, so extending the booking would have required that step. The author's main takeaway is that the task, not the method of control, is what customers care about, and that a full clean matters more than the hourly rate. At $30 per robot-hour, the robots undercut the roughly $40 to $44 per worker-hour that human cleaners charge in his neighborhood, yet the booking still ended incomplete.
The more interesting part of the piece is the economics around that hour. Tau's $30 per billed robot-hour covers a long chain of human work: a staff member dropping off and picking up the robots, help with their shoes, one remote operator per robot, a driver in a van, a person answering text requests, and battery and equipment cleaning between visits. The author argues that renting robots by the task or hour is a much smaller commitment than the $449 monthly plan from Weave's Isaac 1 or the $20,000 purchase price of 1X's NEO, and that shared fleets could raise utilization and generate data that improves performance. He points to apartment buildings or student housing as an efficient setting, where robots could move between units with a battery-swap station on site, cutting driver costs and giving them repeated practice in similar layouts. Tau's co-founder Cornelia Weinzierl told him the robots can already fold many clothing types and load and unload dishwashers, though slowly, and that autonomy would grow over time.
The author closes his analysis by sketching what a larger fleet would require beyond the robots themselves: neighborhood depots for charging and repairs, clear identification of which company sent a robot, safety standards and insurance covering damage to a home or a person, booking ratings, and controls over who can operate a robot and access its recordings. Customers should be able to know what gets recorded and opt out, and the author notes that he accepted data collection for this visit, though other users may not. He would book again, but only if the robots finish the whole house, and he would like a two-hour slot to see them fold laundry.