@ChickenJoeSTL@hallegilb Love that! Gorgeous couple. My great grandparents had a large home on Pershing Place in the 1930s-60s that my grandfather grew up in, and he met my grandmother in the late 50s at a formal dinner at the Chase. I loved them so much, and always stay there when I visit St. Louis. <3
I was interviewed on @PIX11News news yesterday and was on TV last night! https://t.co/UAp6wGLsRy one of my favorite spots in Brooklyn ... Puppetworks, sorta robotics-adjacent, is looking for a new home. FF to 1:30 for my bit 🤪
I have been playing with various natural-language-to-CAD tools integrated with Claude this past week. They have reached an interesting threshold now that you can see that they are capable of generating geometrically detailed models. It isn't perfect zero-shot generation yet but it feels remarkable that you can just take an image of your setup in the real world and ask a VLM to generate a very detailed prompt and then run it through a natural-language-to-CAD skill and it produces something surprisingly good.
Most of my runs were with @GeneralistAI demo screenshots and it took me only 2-3 attempts to get a decent looking model (they are not simready but still worth looking at). All with /render skill from @mfranz_on's repo https://t.co/yiua8K1LdZ that integrates build123 with claude.
Everyone's freaking out about vibe coding. In the holiday spirit, allow me to share my anxiety on the wild west of robotics. 3 lessons I learned in 2025.
1. Hardware is ahead of software, but hardware reliability severely limits software iteration speed.
We've seen exquisite engineering arts like Optimus, e-Atlas, Figure, Neo, G1, etc. Our best AI has not squeezed all the juice out of these frontier hardware. The body is more capable than what the brain can command. Yet babysitting these robots demands an entire operation team. Unlike humans, robots don't heal from bruises. Overheating, broken motors, bizarre firmware issues haunt us daily. Mistakes are irreversible and unforgiving.
My patience was the only thing that scaled.
2. Benchmarking is still an epic disaster in robotics.
LLM normies thought MMLU & SWE-Bench are common sense. Hold your 🍺 for robotics. No one agrees on anything: hardware platform, task definition, scoring rubrics, simulator, or real world setups. Everyone is SOTA, by definition, on the benchmark they define on the fly for each news announcement. Everyone cherry-picks the nicest looking demo out of 100 retries.
We gotta do better as a field in 2026 and stop treating reproducibility and scientific discipline as second-class citizens.
3. VLM-based VLA feels wrong.
VLA stands for "vision-language-action" model and has been the dominant approach for robot brains. Recipe is simple: take a pretrained VLM checkpoint and graft an action module on top. But if you think about it, VLMs are hyper-optimized to hill-climb benchmarks like visual question answering. This implies two problems: (1) most parameters in VLMs are for language & knowledge, not for physics; (2) visual encoders are actively tuned to *discard* low-level details, because Q&A only requires high-level understanding. But minute details matter a lot for dexterity.
There's no reason for VLA's performance to scale as VLM parameters scale. Pretraining is misaligned. Video world model seems to be a much better pretraining objective for robot policy. I'm betting big on it.
@kaseyklimes Communities and families are the foundation of society; the housing crises is a great threat to it and takes priority over tech hiring concerns.
@kaseyklimes Useful anecdote: if you've been inside the housing stock of bed stuy and you have a heart, it would break it to see how many gorgeous homes have been cheaply exploited for airbnb, only to be flipped a few years down the road in an unlivable condition AND configuration.
@Michael_J_Black@ICCVConference Congratulations! I am so grateful to have had this model to use in my research; I can't overstate how much it helped. Well deserved.
@JamesMartinSJ yes! My religion teacher in Jesuit high showed this to me when I was 16 and I remember feeling so captivated, I could not wait to get to that class to catch the next 30 mins of it each day. pure human heart, pure gold.
@ryangrim this is so annoying. cuomo's biggest selling point is being practical leader and this is a grifty thing to propose. i want to like him... but ugh.. its not even a good policy, zero sum BS. leave the current units alone, create more, enforce existing law better, everyone wins
Stretch 3 went on an adventure to the Portland Art Museum! Thanks to Stretch’s moving tablet and mobility, our friend Henry was able to explore the museum like he was there in person. Big thanks to the museum’s accessibility team for making it such a welcoming and meaningful day!