James Erwin

Beginning my new mission to bring humanoid robots to the trades. Will do my best to tweet my progress

@Em_Nomadic x.com/clairemao78/st…

@Em_Nomadic You say ‘teleportation’ a few times in this video but I am seeing many other people using ‘teleoporation’ or a variation of tele-operate. Honest question. What is the correct term?

Robotics Fundamentals: The Embodiment Gap One robot learns to open a door. Another can’t just copy it. Different body, different joints, different limits. The next unlock in physical AI: knowledge that moves between bodies.

@Em_Nomadic And I love this video btw

@PrimeErwin What types of tasks are you looking to start your teleop operators training on first?

We just bought 50 Meta Quest and Pico VR headsets for our teleop operators around the world. Building a qualified global workforce, gamifying their experience and nailing the distribution ahead of massive demand that’s coming is the real moat.

This is what I’m talking about!!!

most people dont have a clue how tremendous the technological shift will be in just the next 3 years.

Had a cabinet sub no show today because of his wife’s birthday. They were supposed to be done so the flooring guys could come through. While I’m trying to keep timelines tight, the cabinet guy doesn’t realize he just made 2 guys not have work tomorrow

Robotics Starts Here is live. 13 parts. Complete beginner's guide. No gatekeeping, no prerequisites, just clarity on what robotics actually takes and where to start building. From understanding the stack to your first robot at Tnkr.ai

Self driving humanoids will have a far bigger impact than self driving cars.

I moved 12 tons of gravel from the street to my front yard via shovel and wheelbarrow this last weekend by myself 24000 lbs of the same monotonous motions over around 9.5 hrs Somewhere around 120 wheelbarrows full There is no way a humanoid robot can’t be trained to do that

This is a succinct argument for humanoid robots

@PrimeErwin @JasonrShuman I have been looking into VLA for training robots for the trades. Devices like the Ego1 by @gi_labs and the Instacore by @instawork I think teleoperation combined with VLA training will help bring robots to construction trades

@JTEIII @JasonrShuman I dont have much information about construction trades but I would guess it's not very easy to navigate and manipulate. Thus, at the beginning, it's 100% teleoperation thing.

Rumor has it that one of the fastest growing robotics players just lost a monster contract/customer after their robot failed to execute the task at the quality bar required. This leads me to one of the biggest questions no one is talking about in robotics publicly. How strong will the switching costs be? Humans get hired when businesses believe they can do a job well. They get fired when they don’t do the job well, lack consistency or someone comes along who can clearly do it 10X better. Robotics is not software so our mental models need to evolve along with it.

@PrimeErwin @JasonrShuman I want to help incorporate humanoid robots into the construction trades. Do you think this will only be possible in the near future with teleoperation?

@JasonrShuman Regionally distributed trained teleoperation workforces is the answer here.

@N8RealEstate A billboard?

Working a deal with 4 vacant residential lots at the end of a dead end street I'd probably just flip the house, not sure what a good use would be for the land Not really a good area to build Any ideas?

I can’t remember the last time I was excited about an album coming out. They did so well making it feel like Sublime with Bradley

The bottleneck of frontier robotics isn’t compute, labeling, or the models themselves. It’s data collection. While language models scaled effortlessly on open internet text, robotics requires physical trajectories, motor torques, and tactile forces that cannot simply be scraped from a webpage. Every token has to be fought for. Here is a breakdown of the 7 data types shaping the industry today, each representing a trade-off between collection cost and action-label purity: 1. Real Teleoperation (AgiBot World, DROID). Collected by humans guiding hardware, it scales linearly with human hours. 2. Low-cost Capture (Mobile ALOHA, UMI handheld). It drives collection cost down while keeping real physics, though it introduces an embodiment mapping problem when transferring human hand actions to robotic joints. 3. Fleet / Deployment Data (Tesla Optimus, Figure). These are trajectories from robots already working in the field. Tesla is betting its automotive fleet infrastructure transfers to Optimus. It generates powerful, real edge cases, but requires scaled deployment. 4. Simulation (NVIDIA Isaac Sim, Genesis). While offering near-infinite scale, the sim-to-real gap still struggles to model contact-rich dynamics like slipping, twisting, friction. 5. World-Model Synthetic (NVIDIA Cosmos 3). NVIDIA just shipped Cosmos 3, which natively outputs action trajectories, not only video pixels. If a world model can accurately simulate the laws of physics natively, it reduces the need for manual teleop data drastically. 6. Egocentric video (Ego4D, Meta’s Project Aria). First-person human video captured with head-mounted rigs. Far more scalable than teleop and closer to a robot’s own viewpoint. Still carries no robot action signals on its own. 7. Internet video (Youtube, TikTok). Maximum scale, lowest cost, effectively free. It captures the widest range of objects, tasks and physical situations, but with zero action labels and (mostly) a third-person viewpoint. Collecting data is only the step one. The next great execution challenge is engineering a coherent training recipe that can blend these heterogeneous data sources into a single model.

@ryanmhickman This is so important for bringing robotics to the trades. The future of construction will be totally different than today

Everyone in Embodied AI is talking about Vision-Language-Action (VLA) models. Almost no one is talking about the physical nightmare of collecting the data to train them. You can't scrape a kitchen table or a warehouse shelf from a web browser. To get to millions of hours of diverse, real-world manipulation data, you need hundreds of rigs. But you can't buy them. If you build them out of off-the-shelf developer kits, they weigh 15 pounds, overheat in an hour, require bulky cabling, and break the first time an operator wears them on a job site. At the Instawork Robotics Lab (IRL), we had to build our own. Meet the Instacore: a rugged, 4lb wearable egocentric data-capture engine designed specifically to survive a full shift on a standard power pack. We didn't build a flashy humanoid. We did hard, blue-collar systems engineering: 💾 THE COMPUTE — A custom MediaTek Genio carrier board that runs completely fanless, routing 5 camera streams directly to on-board storage. ⚡ THE I/O PIPELINE — We ditched USB for industrial GMSL. Thin, ultra-flexible coaxial lines route high-speed data down to the backpack and Power-over-Coax (PoC) back up, completely eliminating batteries on the wrist. ⏱️ UNIFIED SENSOR CLOCK — The MediaTek Genio SoC drives a shared master clock straight to the ISPs driving our 5 global shutter sensors, stamping metadata at the microsecond of capture to ensure zero-drift temporal alignment. 👁️ OPTIMIZED OPTICS — 95 DFOV lenses on flexible PCB ribbon modules keep the wrist cams flat to prevent snags. A 145-degree chest camera captures the macro workspace, while a 50mm baseline rectilinear stereo head pair preserves close-up 3D mapping. To build hundreds of these, we took over a warehouse in Mountain View in April, called it the Instalab, and brought in talented Pros with assembly backgrounds from Tesla, Apple, and Meta. To test the systems, we had our Pros wear active rigs while assembling more rigs. The video below shows the raw, time-synchronized Foxglove playback of that exact loop. Now, we’re shipping these units globally to scale data collection for real Pros on the job.

Try Ego1: - Synced exposure - Hardware synced clocks - Per device calibration

@shumochu I would be hard pressed to get to Austria right now but I would love to try it out

I bring Ego1 with me at ICRA. DM me if you want to try it in person.

2027 is the year of robotics