The Okay-Bot open-source humanoid robots. | Credit score: Okay-Scale Labs
Editor’s Word: Rui Xu is the previous chief working officer of K-Scale Labs, a San Francisco-based startup that attempted to construct low-cost humanoid robots. The corporate shut down in late 2025 and just lately open-sourced its mental property. Xu first printed this text on LinkedIn. It was reprinted together with his permission.
I spent a yr as COO of a YC-backed robotics startup making an attempt to construct reasonably priced humanoid robots. I used to be forty, had 15 years of {hardware} expertise transport merchandise at Intel, Xiaomi, Lenovo, Amazon and ByteDance, and joined to run provide chain and product operations.
The corporate didn’t make it. We by no means closed our Collection A. By late 2025, it was over.
I’ve written about the good parts earlier than. The hackathons, the storage vitality, the primary time the robotic walked. This time I need to write down what I really realized. A few of these are industry-wide traps. Some we walked into ourselves.
1. Giant Mannequin Chauvinism Will Get Somebody Harm
There’s this perception going round that AI fashions are getting so good that {hardware} can afford to be dumb. Sensors? The mannequin will determine it out from imaginative and prescient. Security limits? The coverage will study to keep away from them.
I name this Giant Mannequin Chauvinism. At our startup, it formed choices always. And to be truthful, it wasn’t one individual’s blind spot — most of us purchased into it to a point. The AI was genuinely spectacular, and it was simple to let that pleasure paper over {hardware} fundamentals.
One instance that also bugs me. We spent a genuinely very long time debating whether or not so as to add finish stops to the robotic’s joints. Finish stops. Mechanical restrict switches. A chunk of steel that bodily prevents a joint from destroying itself. Most elementary security redundancy there may be.
The argument towards: the AI coverage ought to study the joint limits. Finish stops are further value, further weight.
Anybody who’s performed {hardware} is aware of why this doesn’t maintain up. Finish stops exist as a result of software program fails. Fashions glitch. Insurance policies hit edge circumstances no person anticipated. When a language mannequin hallucinates, you get a improper reply. When an actuator blows previous its joint restrict at full torque as a result of the coverage had one dangerous inference step, you get a damaged machine. Or a damaged individual.
The mannequin could be proper 99.99% of the time. The top cease is for the 0.01%. Within the bodily world, 0.01% is the one quantity that issues. Even Tesla, with all its autonomy ambitions, nonetheless places brakes on the automobile.
2. Cease utilizing over-simplified analogies. They’re for fundraising, not for constructing.
Each robotics pitch deck has one. “We’re doing for robots what Tesla did for EVs.” “That is the iPhone second for embodied AI.” At our firm, the favourite was the hoverboard (平衡车). Humanoid robots would comply with the identical value curve as self-balancing scooters: costly novelty → Shenzhen mass manufacturing → commodity {hardware} → in all places.
A hoverboard motor simply must spin. That’s it. A humanoid robotic’s actuators should be terribly exact, explosively highly effective, proof against put on, and constant unit to unit. One actuator barely out of spec and the robotic walks improper, or falls. Hoverboards, smartphones, no matter analogy you choose, none of them inform you something helpful about constructing a humanoid.
However “it’ll be like a hoverboard” is a narrative VCs get. Inevitable value discount, China manufacturing magic, billion-unit scale. Each hour spent on analogy debates was an hour not spent on the precise technical issues.
Analogies are compression algorithms. They make complicated issues easy by throwing away info. In a pitch deck, superb. In an engineering determination, the thrown-away info is normally the half that kills you.
3. {Hardware} provide chain just isn’t a job
A couple of software program founders suppose provide chain is a job. Discover somebody who speaks Chinese language, level them at a manufacturing facility, examine the field. This is without doubt one of the most typical methods {hardware} startups get into bother.
Once I joined, there was nothing. No producer relationships, no cost phrases, no QC course of, no logistics pipeline. Constructing it meant meeting, elements, actuators, a number of Chinese language CMs for fabrication. Each meant separate negotiations on pricing, high quality requirements, MOQs, manufacturing scheduling, throughout currencies, time zones, and enterprise cultures that function on utterly completely different assumptions about how offers work.
That’s not “speaking to suppliers.” Manufacturing just isn’t a service you purchase. It’s a functionality you construct. Your relationship together with your CM determines whether or not actuators are available inside tolerance or 2mm off. Whether or not unit value lands at $800 or $2,400. If an organization’s {hardware} operations will be summarized in a single sentence, it doesn’t have a {hardware} technique. It has a hope.
4. There isn’t a such factor as “commodity” in robotics {hardware}
One of the harmful concepts going round: robotic {hardware} will turn out to be “commodity,” assembled from off-the-shelf components by Chinese language producers similar to telephones, with the true worth sitting within the AI layer.
Not but. Not even shut. There’s no customary BOM for a humanoid. No off-the-shelf actuators that simply work for strolling. Each crew constructing a legged robotic proper now could be designing customized {hardware}.
However when an organization buys into the “{hardware} is commodity” story, the injury is actual. The folks constructing the bodily product find yourself with much less voice and fewer recognition than what they really contributed. Energy shifts to whichever perform will get the “defensible” label, irrespective of who’s doing the toughest work.
There’s a sample I noticed rather a lot. I name it Schrödinger’s experience. When one thing goes improper on the {hardware} aspect, instantly they’re not a {hardware} individual, they do not know. When the engineering crew says a redesign takes 4 months, instantly it needs to be performed in 4 weeks. You may’t have it each methods, and the engineers who’re really doing the work can see proper via it.
Our engineers constructed a robotic that walked. That was the toughest engineering the corporate did.
5. In a race, dangerous R&D choices kill quicker than dangerous luck
Everybody in robotics is racing. Capital is there, expertise is flooding in, the market is paying consideration. However a race rewards velocity, and velocity just isn’t effort. Velocity comes from making the best calls quick.
The only largest mistake I noticed was getting caught on locomotion. Months burned, the robotic nonetheless wasn’t strolling proper, and in the meantime the fundraising window closed and rivals shipped demos. This wasn’t only a management name — the entire crew, myself included, underestimated how arduous the issue was and the way lengthy it could take. The GitHub was filled with repos. From the skin it regarded like velocity. From the within it was movement with out convergence. Repos don’t ship. Demos ship. Merchandise ship.
The deeper situation was determination high quality. Impulsive calls kill simply as quick as sluggish ones. Committing arduous to the improper route doesn’t save time, it prices double, since you nonetheless need to undo it later.
R&D velocity isn’t repos or commits or hours logged. It’s how briskly you converge on one thing that truly works.
6. 欲速则不达 — the extra you rush, the additional you fall behind
Our timelines have been a operating joke. It was at all times the robotic walks subsequent week. Each week.
When that’s the tradition, folks begin reducing corners to hit unimaginable dates. Engineers write code with AI instruments with out reviewing it correctly. Sensors go onto the product with out calibration. After which the demo fails, once more, and the timeline resets to subsequent week.
That is what the Chinese language name 欲速则不达. Actually: need velocity, fail to reach. When unrealistic deadlines turn out to be the norm, the crew doesn’t really transfer quicker. They simply skip the steps that make issues work. And each skipped step comes again as a failure that prices extra time than the shortcut saved.
The injury goes past engineering. Whenever you make guarantees to your contract producer based mostly on fantasy timelines, you burn the connection. A CM wants life like expectations to plan their very own manufacturing. A chaotic “transfer quick, break issues” mindset would possibly work in software program. It doesn’t work when your producer is allocating manufacturing facility flooring time based mostly on commitments you’ll be able to’t hold.
A private observe
I may have been a greater COO. Ought to have been firmer earlier concerning the organizational issues, again after they have been fixable. Ought to have pushed more durable on life like timelines as an alternative of letting them slide. That’s on me. However I realized the place these traces at the moment are, and that’s one thing I take into no matter comes subsequent.
However I used to be there for the entire thing. First hackathon to final provider electronic mail.
In the event you’re a younger engineer at a startup: belief your instincts on physics. If the mathematics says the joint will break, write it down. Make the case formally. Don’t let the strain to maneuver quick bully you into ignoring what you already know to be true. Your repute is constructed on what you ship, not what you promised.
If these six classes assist somebody, a {hardware} founder, a provide chain individual, a forty-year-old dad or mum questioning whether or not to affix a startup, then this was value writing.
I nonetheless consider in embodied AI. I simply consider it deserves {hardware} that’s as significantly engineered because the software program controlling it.
Concerning the Writer
Rui Xu is a {hardware} veteran based mostly in Silicon Valley. He beforehand served as COO of Okay-Scale Labs, a YC-backed robotics startup constructing reasonably priced humanoid robots. Earlier than that, he spent 18 years transport shopper {hardware} at Intel, Xiaomi, Lenovo, Amazon, and ByteDance, together with merchandise just like the Xiaomi Mi Field, Lenovo Sensible Show, and Amazon Hearth TV. He writes about robotics, {hardware}, and the truth of constructing bodily merchandise at ruixu.us.
The put up 6 classes I realized watching a robotics startup die from the within appeared first on The Robotic Report.
