Tom Jacobs

@tonyzzhao @ericjang11 currently reading a.co/d/0gzfM2qy

Sad but true

@antoinemarcel cool

Funny to be one of the first non-wizards to have a functional, low-cost, general-purpose robot at home

@jhong 100%

Elon always makes big claims that he will do impossible things in x years. His haters relish in things taking twice as long so they can call him full of shit. But if he had said he could do those things in 2x years from the getgo, they would still have been unbelievable accomplishments. I suspect he subconsciously (but maybe even intentionally) offers optimistic timelines because it pushes himself and the companies to work faster.

@Scobleizer @BerntBornich @karpathy @1x_tech Good to see you guys at the launch! @karpathy hug robot now

I just interviewed @karpathy for the first time in my life at @1x_tech. Right after he got a demo. I will have his video up later but turn up the audio and listen to what I have learned. The event here is sick. A robot checked me in. In my life I went from building Apple IIs to this. What a time to be alive!

@GoingBallistic5 I saw the H2 as well, are three prices accurate?

Waisted Opportunity I forgot to mention in the video that Unitree pivoted from a serial waist in the prototype to a parallel kidney waist for production a la G1 Packaging and performance must have been better

GTC 2026. These robots are better.

GTC 2026. The robots are so slow.

@benjamin_bolte instagram.com/reel/DV78paeAT…

Haidilao in Cupertino has an AGIBot X2

@ChongZitaZhang wow, it begins to be over for human athletic ability

zzk273.github.io/LATENT/ really cool work, humanoid playing tennis against humans

@travisk Summary in order: x.com/tompjacobs/sta…

Atoms. atoms.co/vision

@paulcjh_ ow!!

I also now have permanent nerve damage in my thumb from almost cutting it off in my cnc so that's fun

I've been hard at work the past few months in SF working on some buy buttons. Today I'm launching my plug-and-play product line that lets you build a robot real fast. Right now to make a robot you have to stitch together a bunch of different PCBs with jumper cables and wait weeks for blackbox actuators to arrive from China. You spend lots of time debugging why your CAN bus isn’t working, why every actuator performs differently, and meanwhile your wires keep coming loose. So, I'm making the following: - RB1: A robot main board powered by an Nvidia Jetson. This handles power distribution, compute, and a bunch more. - WM1: 2-channel wireless radio for sending video/data making the RB1 remotely controllable over USB-C. - M1: A pancake BLDC motor machined in-house. - ACB3: An FOC control board with matching connectors to the RB1. (big brother to ACB v2.0) - A1/A1m: A planetary/cycloidal actuator powered by the ACB3. (this is on the site in a couple weeks) Everything is on sale for this week, and shipping begins this spring! As a big thank you to the supporters of ACB v2.0 (and thanks for patience in shipping delays) you can buy the ACB3 for 50% off. If you like this kind of thing and want to join please DM me, I'm working solo right now and need good folks to join!

My friend made CADable... prompt your own Onshape CAD.

@chr1sa @JacobZietek @lukas_m_ziegler thanks chris! we met years ago at self racing cars. you helped me fix my RC self racing car, thanks!

@TomPJacobs @JacobZietek @lukas_m_ziegler $3k YAM – I2RT Robotics share.google/zCMwtRIIvlgRKP…

so impressed the I2RT YAM— less than an hour from unboxing to scripting and everything just works the barrier to entry for robotics is lower than ever

@prime_cai nice good team?

@rhoda_ai_ i like it, can i see?

03.10.26

To understand what it takes to build a humanoid robot with model-based control, we finetuned @physical_int 's (PI) Pi05 model for our custom use case and environment. We incurred ~$10K in hardware costs, compared to the typical ~$20K set up (DROID/ALOHA). Here are the lessons and challenges we faced building the first working prototype (shown in the video) in 3 months. Part 1: Hardware, Software, Model Selection, Custom Embodiment, Inference, Embedded Hardware, Hierarchical Planner Part 2: Model Evaluation, Data Collection, Model Training, Simulation and Teleoperation We hope sharing our experience accelerates the learning of others who are in a similar starting point.