Amazon has quietly stood up a Personal Robotics Group dedicated to building a new consumer home robot, according to job postings reviewed on November 8. The team is hiring a Field Support Engineer for customer repairs and a Principal UX Designer tasked with defining, in the company's own words, "what it feels like to live with a Fauna robot."
The group is built around Fauna Robotics, the New York startup Amazon quietly acquired in March 2026 along with its roughly 50 employees, including founders Rob Cochran and Josh Merel. Fauna's product, Sprout, is a 3.5-foot humanoid with a soft exterior, minimal pinch points and expressive yellow eyebrows, designed to be small and gentle enough to share a room with children and pets.
Why Amazon is trying again
This is Amazon's second serious run at a home robot. Astro, the company's wheeled assistant, launched at $1,450 in 2021 and shipped to only a few thousand invite-only customers before stalling out as what one long-time owner described as a glorified trip hazard good mainly for checking on pets. Former consumer robotics VP Ken Washington has said leadership debated giving Astro arms but rejected a bipedal design, even as Jeff Bezos reportedly voiced disappointment with the robot's limited abilities.
That history shapes the new effort. Amazon staffed it with much of the original Astro team, added an MIT roboticist, and pulled in engineers from Lab126, the hardware unit behind Kindle, Echo and Fire devices. One job posting sums up the mandate bluntly: "make robots people actually want to live and interact with in everyday human spaces." Neither price nor a release date has been disclosed.
The crowded race for the living room
Amazon is late to a field that has gotten far more competitive since Astro's debut. Tesla, Figure, 1X and a wave of Chinese manufacturers are all chasing general-purpose humanoids, though most are aimed at factories and warehouses first because industrial customers pay sooner and tolerate rougher edges than households do. Sprout's small stature is a deliberate hedge against that risk: a robot that tips over at 3.5 feet tall is far less dangerous, and far less expensive to build, than a full-size humanoid.
The trade-off is capability. A smaller, softer robot cannot lift heavy objects or reach high shelves, narrowing Sprout toward chores like picking up toys or fetching snacks from a low pantry. Whether that is enough to succeed where Astro failed depends on execution Amazon has not yet shown, and on a price point it has not yet set. For a company that walked away from buying iRobot in 2024 over antitrust concerns, rebuilding a home-robotics push from an acquired startup is a lower-risk, slower path back into customers' living rooms.