2/ The barrier to entry in robotics is incredibly high. Developers / companies build their own hardware, or buy off-the-shelf, and proceed to spend MONTHS building things that have already been built before. Navigation. Teleop. Basic pick and place. Exploration. Spatial

3/ Companes do this themselves because any open source implementations exist somewhere in the github ether and haven't been touched or maintained in 5 years.

4/ Developers are already running our open-source OS and SDK on OEM robots, and accessing the skills or "apps" like this one with a single runfile.

5/ As hardware cost drops to <$2k for dogs and <$5k for humanoids, pumped to see what y'all build! Comment below what embodiment you're shipping on so we can prioritize your integration ASAP.

6/ Join the discord here for early access! (Releasing as soon as we get our shit together and write documentation) https://discord.gg/uSH7nZtS4

@stash_pomichter DARPA already has this

@Steventester44 I'd hope so. But developers don't :)

@stash_pomichter 🔥🔥

@stash_pomichter Lfg 🤩

@stash_pomichter 🙋

@stash_pomichter so cool for 3d realtime , is that lidar top there

@stash_pomichter Any drift ? For uAV ? Wow impressing af

@stash_pomichter Good job,And how to get started?

@stash_pomichter 激光雷达需要什么标准的呀？集

@stash_pomichter I think you guys need to look into @AukiNetwork. They can help you with spatial data.

Tesla - collects 4.3M hours of driving data - every day - for free - to train a 2DoF system (steering + throttle). - yet full autonomy remains unsolved. Frontier robotics startups/labs - collect or purchase 0.01M–1M hours of data - every X month - for millions of dollars - to

Should humanoids have tails? Current legged robots struggle with center of mass control and quick stabilization. Adding more DOF to legs gets complicated fast. Nature already solved this: Cheetahs, kangaroos, dinosaurs – they use tails for dynamic balance, sharp turns, and

Meta's Chief AI Scientist Yann LeCun offers a critical take on the humanoid robot boom. Speaking at MIT, LeCun claimed the "big secret" of the industry is that current companies "have no idea" how to make their robots "smart enough to be generally useful." He argues that while

I built a biologically inspired spiking neural network from scratch and it learned with %5 accuracy to do addition :) There is no backpropagation, no artificial loss functions - just spikes, synapses, and dopamine-like reward signals. it uses STDP -> "Spike-Timing-Dependent