Nilli Mtz

Spent the last few months talking to warehouse and logistics companies and the robot OEMs that serve them. Everyone's excited for Physical AI. But the gap between what we show and what's acceptable for a daily task on a customer's site? Still wide.

Walden Robotics went out of stealth today. The videos are impressive. The question is whether they survive the real world. What they're doing differently: focus on unit economics. Tedrake said: "success is not assured." That humbleness is what makes robots provide value.

@lukas_m_ziegler you are so right, and there are so many more areas that usually go unrecognized but they are so key to the process: supply chain, field support, deployment engineers, they also have to understand how the systems interact to do their job

It's good to see robotics finally getting the recognition it deserves these days. The venture capital market says so, the steady flow of talent says so, and of course so does the progress in AI, which promises that the physical world can be automated, despite being so unpredictable and messy for machines. That's exactly why I wanted to shine a light on the people building in the robotics space. They're a bit like Renaissance men. Their skills go way beyond writing code. They have to bridge software with the real world, and despite how it might sound, that's an incredibly hard challenge. Atoms are harder to control than bits. Here are just some of the fields a roboticist must be aware of: → Software engineering & robotics frameworks: Fluency in Python/C++ and tools like ROS 2 and simulators. Modern robots are software-defined, so code quality is critical. → AI & ML for perception and control + data: Computer vision, sensor fusion, and deep/reinforcement learning. Knowing when to use classical vs. learned methods is key. → Systems integration & mechatronics: Understanding how sensors, actuators, electronics, and embedded systems fit together. Debugging across the hardware/software boundary sets strong engineers apart. → Control theory & motion planning: From PID and MPC to trajectory optimization and path planning. This is what makes robots move reliably in the real world. → Collaboration & safety-aware design: Communicating across disciplines and designing for reliability and safety standards. Deployment realities and failure modes matter as much as features. Hats off to all these great people working in robotics space, and focusing on really hard problems :) @jbhuang0604 deserves an award for that meme :) ~~ ♻️ Join the weekly robotics newsletter, and never miss any news → ziegler.substack.com

@RussTedrake congratulations to the team! the blog and the video bring so much optimism to the space, especially seeing the robots being deployed in factory settings first. I'll keep an eye on future posts to see how different they are from Agility.

In January, I started "building something new" with an incredible team. Today I finally get to share some first details about what we've been building. We've called it Walden Robotics (waldenrobotics.com). I thought long and hard about my own reasons for starting this company. It's not only about the robots. It's also about people. I've tried to capture those thoughts in my first Walden blog post: waldenrobotics.com/news/why-walden It's been an incredible ride so far. Within just a few months of forming the company, we were already operating a general-purpose robot with an end-to-end policy in production in one of the most important factories in North America. It's amazing at how much I've already learned from that experience. There is a lot of work to do, but the mission has never been so clear. Please help me welcome Walden Robotics into the world. And stay tuned for more updates! youtube.com/watch?v=fewvZr…

@CyberRobooo the videos are impressive, I wonder if they are also working on an orchestration algorithm to sync multiple robots and achieve more complex tasks

Xiaomi Humanoid Robot New Factory Training Since Q1, Xiaomi’s humanoid robots have been training on dual-side operations at the self-tapping screw workstation in its auto factory, achieving a 98% success rate. Today, Xiaomi shared that the CyberOne Gen2 autonomously performed two repetitive tasks in the final assembly logistics area: sorting center console side panels and folding material boxes --both reaching 90% success rates. They deployed both bipedal and wheeled models, working together with material transport robots as part of the broader intelligent automation system. I noticed the wheeled CyberOne was clearly more efficient and stable on the panel sorting task. In the first video, you can also see some noticeable jitter when the bipedal robot stops moving (though this should be relatively easy to fix). Xiaomi also candidly pointed out a current limitation in dexterity: when folding the second side of the material box, the robot still needs to adjust its orientation first to face the buckle, whereas skilled workers can pull the ring directly by feel without any micro-adjustments. Real-world scenarios are indeed the best benchmark.

@cchevyyc love this!

how many customer calls in a day is too many? (I love talking to my customers)

@antoinemarcel really cool! I would love to hear about how you are thinking the onboarding process will look like for the first users!

Finally working :) The 1st home robot that's really easy to set up (< 30min for your 1st task), community-driven (if someone teaches it how to grab a beer and bring it to them, it’ll be able to do it too), and low-cost. Shipping 50 units soon. Link below if you want to join the private WhatsApp group.

Legacy robot vendors deploy through System Integrators. Startups want to do it all themselves. Early on that's the right move. You learn the business, you learn the customer. Long run, they are shooting themselves in the foot. Every new site is another hire. Agree? DM me.

If you know any robot company seeing this level of growth in Europe or US, send me a DM. I'd love to talk.

The sales manager corrected me. There's a better reason that drives all of it: hourly wages are even higher in Europe. So the ROI on automation is greater in that context.

One thing that surprised me talking to a warehouse robotics vendor: They have more deployments in Europe than in the US.

@benjaminhgordon @SymboticTweet @ArmsInnovations This move is closer to how Amazon does things, they are realizing that they can get more done the minute that they own the orchestration layer and beyond so that they have more control over its optimization

Warehousing automation is converging with warehousing software. @SymboticTweet just bought @ArmsInnovations. As a result, Symbotic is combining warehouse execution systems (WES) with warehouse management systems (WMS) to build what they are calling enterprise-level Warehouse Operations Optimization. Smart supply chain companies will continue to evolve and combine capabilities to deliver superior innovation. dcvelocity.com/technology/war…

@RoboDaily They are on the right track, however deployment involves multiple steps not only the training part, Standard Bots needs to also think about how to scale their deployment without scaling the amount of operators doing this demonstrations and troubleshooting

Their flagship robot, Core, is already being deployed for machine tending, welding, palletizing, inspection, assembly, and material handling.² What stands out is how easy Standard Bots is trying to make deployment.

American-made robotics is becoming important again. And @standardbots is positioning itself directly in the middle of that shift. And it’s not just a single robot, but an entire AI-native robotics stack designed to modernize manufacturing from the ground up.¹

If you have an opinion about robot deployment, or know someone who does, send me a DM, I'd love to chat.

That's getting in the way of how they scale. As an engineer myself, I get it. But founders need to think about this early enough to build the foundation into their tools.

Spent the last few months talking to warehouse and logistics companies and the robot OEMs that serve them. Everyone's excited for Physical AI. But the gap between what we show and what's acceptable for a daily task on a customer's site? Still wide.