cofounder @Ultraroboticsco. building robots in nyc

Joined August 2019
55 Photos and videos
$450/month is the present, not the future
$450/month to fold laundry? I’m struggling to believe that humanoids will be the future
2
16
976
Very expensive kettlebell
Proof.
1
468
Robots need to be cheap to be abundant
We are finally seeing the launch of cheap, capable robot hardware that is at consumer price levels. new announcements from Weave, Nori Robotics, and Bracketbot show the wave of future consumer robot products that will live in your home and do your chores. more: open.substack.com/pub/itcant…
4
30
2,859
Congrats @kaandogrusoz @evan_wineland on the launch! You guys cooked with this one 🔥 Cool to see how much the robot has evolved and also how much of the DNA has remained the same from the earliest prototypes
Today, we’re launching our home robot Isaac 1. Isaac 1 deliveries will begin this fall. Order yours below.
1
15
1,894
If you include UMI style datasets there is even more. iirc pi0 was trained on ~10k hours so definitely sizable enough to do some interesting work!
I think this means you can collect ~10k hours just from open source datasets, which means basically anyone should be able to build a decent robot foundation model: 500 from the new BitRobot dataset 500 from galaxea huggingface.co/datasets/Open… 3000 from agibot huggingface.co/datasets/agib… ~3000 from Open X embodiment (though it's mostly pretty bad data) robotics-transformer-x.githu… 830 from EgoDex github.com/apple/ml-egodex ~30 from humanoid-everyday (but good quality) huggingface.co/datasets/USC-… 3500 from ABC huggingface.co/datasets/XDOF…
2
18
2,764
who's making buff robots?
35
29
350
16,938
so inspirational If this works it will radically change how we think about preventative healthcare There are still so many things out there to build and make the world a better place
A technical dive inside our new "Midjourney Scanner"
1
4
614
relentless hill climbing can result in some pretty cool stuff 👀 sim2real in <60s next?
ultra fast RL achieved! g1 policy trained for < 60s at 1.3M SPS on a single rtx pro 6000. all open source! nanog1.com
2
1
13
2,607
pls stop with the robot ai slop 😭
The quiet trick in robot teleop: pilots wear VR headsets not for immersion, but to be deliberately blinded. This forces data collection under the robot's exact sensory limits, yielding cleaner training than full human perspective. If the human can peek around the robot's camera, the policy learns from info it will never have at runtime. Tight embodiment isn't UX, it's a data integrity constraint.
4
35
4,398
very cool haptic exoskeleton -- i really want to try one i wish they showed some tasks that actually required force feedback though. most of the demonstrated tasks can be done pretty consistently through teleoperation w/o force feedback...
Introducing Universal Manipulation Exoskeleton (UME) A low-cost exoskeleton with real-time haptic torque feedback for learning autonomous policies that perform highly force-mediated, tightly space-constrained, visually occluded, whole-body, and long-horizon mobile manipulation tasks. Using UME, the teleoperator can unsheathe a heavy metal sword completely blindfolded. ume-exo.github.io/ 🧵1/N
5
1
19
2,580
In robotics, if something seems easy, it's probably too good to be true. There's a lot of juice to squeeze, but it frequently comes from doing the hard thing (even if it's not scalable) and compounding your learnings
opinion piece: cargo cults, data flywheels, and novelty pumps tl;dr: “deployment data flywheels” miss that scale should provide novelty, and deploying robots on a known task wrings most novelty out of its generated data your ops team’s novelty pump is the only way out 🔗👇
3
2
35
4,995
Really cool concept. Should make it much easier for people to get their first robot built and running!
Introducing Kits by Tnkr 📦 The fastest way to clone a robot. For Maintainers: distribute your project, BOM to kit in one click. For Builders: everything you need to build it, in one box. tnkr.ai
2
2
25
4,131
Chetan reposted
Concentration of power, capabilities and economic wealth is the biggest risk in AI. We need open science and open-source more than ever!
115
479
3,106
168,598
Lots of people hate on webrtc, when actually the problem is the implementation of webrtc. There are tons of knobs to tune and most of the defaults suck for robotics. The libwebrtc implementation is particularly rough
Comparing against the WebRTC in chrome is not really apples to apples, it’s tuned for smooth video with a jitter buffer that can easily add 50-100ms. Removing uncontrollable things like network conditions and fixing the jitter buffer, 10ms glass to glass is easily doable over WebRTC. WebRTC’s actual overhead is < 0.5ms. The rest of the latency is camera capture/encode/network/decode/render.
1
8
706
Congrats to @evanbeard and team! NYC robotics running strong 🚀
Today, we’re thrilled to announce our $200M Series C funding round at a $1B valuation, led by @RoboStrategy and existing investors including @generalcatalyst. Standard Bots is now America’s largest manufacturer of AI-native industrial robots. Our customers include Sunoco, Lockheed Martin, NASA, and the US Army along with hundreds of other manufacturers across the country. We’re proud to say that we’re on track to deploy 10% of all U.S. industrial robots by next year. We are expanding our Glen Cove, New York facility to 70,000 square feet to scale our vertically integrated production process. We currently design almost all our own parts, including our own actuators, and we assemble every final product in-house. By 2027, we’ll manufacture everything — from metal in to robots out — right here in America. We believe AI-native robots are the essential power tool of the 21st century — the tool that will grow American manufacturing and help every American worker to be a force at work. You just show your robot how it’s done, and it learns through demonstration. No coding, no consultants, just unbox and deploy faster than anything else on the market. Right now it’s possible for the United States to revitalize our manufacturing base if we become the worldwide leader in this transformative technology. We must build American robots, and put them to work in American factories. It’s a national imperative, and it’s our central mission. This fundraise gets us one step closer to the goal. The future of American manufacturing is bright! Join Standard Bots, and show your robot how it’s done — we’re just getting started.
1
13
2,298
Chetan reposted
This is true. The bots right now are the worst they’ll ever be.
"Dominate Global Robotics" doesn't seem great Good read after @corememory @kyliebytes piece on Westmag, Atlas Motion Systems
1
7
36
4,426
turns out, if you want to deploy robots, making them easier to deploy helps
The advantage of a humanoid like this is speed of deployment
2
25
3,740
policies that don't have memory can sometimes exhibit things that feel like memory (e.g. tracking that it dropped something on the ground) through: (a) fast reaction time active perception (b) encoding memory through body language/posture
In an impressive edge case example, OP1 recognizes when a box falls to the floor and cleanly picks it off the ground and resumes it’s task.
1
33
3,711
having an embodiment that can clean up after itself is massively valuable for customers (who don't want to deal with all the stuff your robot dropped on the floor)
In an impressive edge case example, OP1 recognizes when a box falls to the floor and cleanly picks it off the ground and resumes it’s task.
2
3
33
3,402
Chetan reposted
Autonomy progression from June 2025 → June 2026 One year ago our robot was just scratching the surface of dexterous manipulation, moving a cube to a marked spot on a desk. Today, OP1 autonomously picks builds boxes, and we’re just getting started. Autonomy compounds. The robot running today is the least capable version of itself you'll ever see.
2
9
51
4,023