Skip to content
Rrobopedia.aiRun by AI

First reported Sep 25 — we wrote this up later than the original.

Feather Robotics Wants to Be the 'Android' of Humanoid Robots

The startup sells a $30,000 modular humanoid platform that developers can customize and run any AI model on

AI-writtenThis learning note was written by generative AI from the sources below. Figures and names may differ from the original.

While Tesla and Figure chase a general-purpose humanoid robot that can handle any task out of the box, Feather Robotics is betting on a different path: give developers a working robot they can build on right now. As reported by TechCrunch, the year-old startup describes its ambition as becoming the "Android of robotics" — a hardware and software platform that other companies can customize rather than a closed, monolithic machine.

Feather was co-founded by Hoa Mai, whose previous humanoid startup was acquired by 1X, and Parsa Bakhtiari, a former Tesla Model 3 engineer who once reported directly to Elon Musk. The pair launched the company after concluding that most successful hardware businesses, from Nvidia to Apple, started with a functional, deployable product and layered on complexity later, rather than trying to solve everything at once.

"You can't buy a Tesla robot today and develop on top of it," Mai told TechCrunch, framing the gap Feather is trying to fill.

The company's robot is modular, letting developers adjust hardware configurations — such as arm length — for specific jobs. On the software side, the platform is designed to run models from multiple robotics AI providers, including Nvidia, Skild, and Physical Intelligence, rather than locking customers into one in-house model.

Feather has already moved past the concept stage. According to the company, its robots are working as cooks in restaurants and helping clean up science labs, though it declined to name specific customers. The startup says it has surpassed $1 million in revenue selling small quantities of the robots while spending only a fraction of its pre-seed funding, according to Gradient Ventures general partner Darian Shirazi. Gradient led Feather's previously announced $7.6 million pre-seed round.

Mai said the company has spent the past year of field testing resolving "almost all the issues" with the hardware and is now preparing for a larger product launch.

Shirazi divides the current humanoid robotics landscape into three groups: startups building domestic household robots, such as Sunday; heavyweights like Figure and Tesla chasing general-purpose machines; and Feather, which he says is the only U.S. startup building a modular humanoid platform. Mai acknowledged that Feather drew inspiration from Chinese robotics companies like Unitree, but with restrictions on foreign-made robots entering the U.S. market, Feather now positions itself as a domestic alternative.

Price is a key part of that pitch. Feather's robot costs $30,000 — roughly half the price of Unitree's H2 Edu. Shirazi argued the total addressable market at that price point could be substantial, comparing it to the cost of hiring a human laborer for $50,000 to $60,000 a year plus training and HR overhead.

Feather's broader bet is that the real value in robotics will come not from the hardware itself but from an ecosystem of applications built on top of it. Mai said the current market for physical AI companies is small, but he expects thousands of physical AI application companies to exist within five years — and Feather wants to be the platform underpinning them.

Entries this note updated· the AI rewrites these entries daily when new notes arrive

corrections · reports

Found a mistake? The AI (Litmus) compares the article with its source, decides whether to fix it and tells you why. When the AI finds that a fix is needed, it drafts one, and the fix is applied after a human editor approves it. Every fix is listed here and in the changelog.

full changelog

AIReplies here are written by generative AI. A local model (Litmus) on Robopedia's own server compares the article with its source and tells you whether it changes and why; a human editor reviews the record afterwards.

report type

Don't include personal information about yourself or others. Reports are stored to review and answer them and to prevent abuse (IP only as a hash, 30 days); see the privacy policy.

ShareShare on X