Mixed-case palletizing — building a single pallet out of dozens of different products — is one of the hardest jobs in a warehouse to automate and one of the most miserable to do by hand. On September 10, Maven Robotics, a Santa Clara startup founded by two brothers out of Apple’s special projects group, exited stealth with a $100 million Series A and a robot it says is already doing that job for a Fortune 250 company.
🔍 THE BOTTOM LINE: Maven is entering the most crowded robotics funding market in years with an unusually narrow wedge — not a humanoid, not a platform, but one hated warehouse task done well enough that a large consumer goods company is already running it across multiple daily shifts. Whether that task-first bet scales beyond the demo floor is exactly what the $100 million is supposed to test.
Two brothers from Apple’s special projects group
Hamza Derbas, Maven’s co-founder and CEO, spent about nine years on Apple’s special projects group — widely reported as the company’s self-driving-car unit — before leaving to start Maven with his brother Khalid, the company’s CFO. The launch announced September 10 pairs the Series A — from RoboStrategy, LocalGlobe, Vine Ventures and XTX Ventures — with a team the company says brings more than 200 combined years of experience across Apple, Tesla, Rivian, Cruise, Zoox, Ford and Archer Aviation.
“We’re grounded in solving one customer problem at a time,” Hamza Derbas told TechCrunch ahead of the launch.
That is a pointed contrast with how most of the general-purpose robotics field has pitched itself over the past two years, which has been roughly: build the general body first, find the work later. Maven is arguing the order should be reversed.
The job nobody wants, and a robot built for it
In consumer-goods distribution, a large share of outbound volume ships as mixed pallets, and the mix keeps changing. “Within 48 hours of them putting the stuff on the shelves, they want to change the mix based on real-time demand,” Derbas told TechCrunch. “It’s all done with human labor today, running around the warehouse picking one of this, one of that.” That constant reshuffling is why mixed-case palletizing resists conventional automation: a fixed cell tuned to one product flow breaks the moment the flow changes.
Maven’s machine — visible in the company’s launch photos as a wheeled pedestal with two yellow arms, closer in shape to a double-arm cobot than to a humanoid — starts with mixed-case palletizing and tote handling, a market the company estimates at roughly $80 billion. From there it plans to expand into material handling and assembly, which it sizes at more than $1 trillion. Both are the company’s own estimates and should be read as such.
What is actually running, and what is promised
The deployment claims are the most concrete part of the launch. According to the company, Maven has fleets running autonomously across multiple shifts daily at a Fortune 250 consumer packaged goods company. TechCrunch’s profile reports eight robots working 16-hour shifts at 99%-plus uptime, with a plan to build 250 third-generation robots. The company is targeting more than 100,000 hours of autonomous real-world operation by the end of 2026 and more than a million hours by the end of 2027.
All of those figures come from Maven and its coverage of Maven, so they deserve the standard discount a launch gives itself. But the shape of the claim is falsifiable in a way most launch-week claims are not: either hundreds of robots get built and the hour counts climb, or they do not.
A different shape in a crowded market
Global robotics venture funding topped $18.8 billion in the first half of 2026, according to figures Maven cites in its launch materials, and most of the headlines have gone to humanoids: Agility is expanding its humanoid deployments across manufacturing, Bosch plans humanoid robots on a production line in Buhl from 2027, and Boston Dynamics’ Atlas now swaps its own battery. Against that backdrop, Maven’s pitch is almost contrarian: the body was never the point, the task is.
The honest read is that both bets are still bets. Humanoid companies have to prove their general machines can do specific work reliably; Maven has to prove a narrow machine can generalize outward without losing the reliability that won it the first customer. The difference between a demo and a deployment is uptime, and 99%-plus across 16-hour shifts — if it holds as the fleet grows past 250 robots — is the number that matters more than any form factor debate.
❓ FAQ
What does Maven Robotics’ robot actually do? According to the company, it performs mixed-case palletizing and tote handling — building mixed pallets from varied products — with planned expansion into material handling and assembly.
Who founded Maven Robotics? Brothers Hamza and Khalid Derbas. Hamza, the CEO, spent about nine years at Apple’s special projects group, widely reported as its self-driving-car unit; the wider team includes alumni of Tesla, Rivian, Cruise, Zoox and Ford.
How much did Maven raise, and from whom? A $100 million Series A announced September 10, 2026, from RoboStrategy, LocalGlobe, Vine Ventures and XTX Ventures, per the launch release.
Why should New Zealand logistics operators care? The labour problem Maven targets — unpredictable, fast-changing mixed-case picking that is hard to staff — is the same one felt by distribution centres here, and if task-specific robots reach deployment maturity, they will appear in local supply chains long before general humanoids do.
🔍 THE BOTTOM LINE
The general-purpose robot race has mostly been argued over bodies: humanoid versus wheeled, dual-arm versus single-arm, legs versus wheels. Maven’s launch argues the argument is premature — pick one task that costs real money, do it reliably at scale, and the generalisation question can wait. If the hour counts and the 250-robot build-out land on schedule, the company will have evidence for the task-first thesis. If they slip, it will be one more stealth story that left stealth early. The next twelve months of uptime data will settle which.
