Shapath Das

Introducing PRM (Physical Reasoning Model): a foundation model that infers and outputs an uncertainty-calibrated physical state from robot experience, so downstream policies can train on physics-complete signals — not just pixels and actions. @unsupervizedai @Ranita_Jana

@babugi28 Congratulations!

Human Archive’s the only YC startup mentioned in the headliner for today’s top news, and our demo day pitch received the #1 investor inbound interest. 🚀🚀🚀

@babugi28 Looking forward!

@narsagna congratulations!

We got skillsync.com

@xeepak @ChatGPTapp Good suggestions Deepak, also these days the answers are taking longer inference time compared to @claudeai

@ChatGPTapp Please allow us to minimise the answer to a prompt, or allows sub-threads. When I ask a broad question, I often have to ask multiple sub-questions to understand the original response. But scrolling gets quite difficult after 2-3 sub-questions.

@after_ephemera @partabot_ Yes, indeed, they delivered the bot a week late.

@shapathdas @partabot_ Did you ever get hold of them? I ordered recently and havent seen any updates come through

@partabot_ Hey what is the best way to get in touch with you?

who wants bounding boxes and segmentations? we’re launching next week, and researchers in academia are eligible to receive our datasets for free. drop a comment and i’ll dm you

This concept bot has so may issues! Why it even has two legs why not wheel based!

@ShubhamKukretii @Visa Really cool! So so happy for you both!

BIRD IS HERE!!! Excited for our partnership with @Visa

@AnselDias @norma_core_dev Super cool!

Just built the ElRobot Manipulator from @norma_core_dev - 7dof Arm using Feetech servos - 1dof Parallel Gripper - Far more stable and precise than SO101 - Body Panels for better cable runs I think this arm can be capable of assembling another one of itself. More soon.

Hi @danfei_xu just finished reading your Article titled "To summon a Sensorimotor Ghost", really insightful, thought I should invite you to read my take on the data problem.

Danfei and the EgoMoma team needs a clear shoutout!

Humans have evolved an intuitive understanding of physical laws—not by learning from scientific books, but through rich, multisensory experience of the world. Now we want machines to acquire a similar understanding purely from video data. Is it working? x.com/shapathdas/sta…

@mattfreed Thank you for sharing @mattfreed

I receive a lot of messages asking where to get parts made fast So I've built an ongoing list of US manufacturers that offer quickturn solutions I decided to only include manufacturers who offer instant quoting - or platforms online that are moving in that direction No emails, no forms. This list is currently small, but my hope is that it helps you find quick-turn partners and reveals where gaps exist: processes still waiting to be brought to quick-turn, and legacy industries yet to unlock instant quoting. You can submit new manufacturers, request new quickturn solutions, and help keep manufacturer information up to date. I hope you can contribute, and I hope this list continues to expand as more quickturn offerings become available Link in replies Pic related

Hey folks, Is anyone traveling to Bangalore from San Francisco this week?

“Strategy is product, and product is strategy” — Jan Simon Veicht

“What once felt magical now felt inefficient. What was left were nostalgic memories of the good old days, which in hindsight weren’t that good. That’s how paradigm shifts work.” A few best lines from the @Alfred_Lin essay!

Robotics isn’t compute-constrained. It’s data-constrained. Specifically: #teleoperation_data.

Compare that to AI LLMs → trillions of tokens (free-ish) Vision → billions of images (internet-scale) Robotics → ~10⁵–10⁶ trajectories (painfully collected) We’re 6–9 orders of magnitude behind.

Teleoperation is to robotics what labeled data was to deep learning -- scarce, expensive, and ultimately the defining bottleneck.