Hardware founder Yacine argues well-funded robotics companies are using the wrong methods to integrate AI with physical systems
The debate pits algorithmic improvements against intensive data collection.
Some users praise the builder's unique AI robotics approach rejecting Silicon Valley methods and emphasizing algorithms over data as the correct path, while others harshly dismiss the claims as amateurish or idiotic.
Most Activity
The fact that robotics isn't unilaterally solved is just an algorithms gap. All of the silicon valley companies doing "robotics" right now with their data collection meme are on the wrong path entirely.
It's surprising that robotics isn't really, actually solved. It's not hard
Probably the real limiter on this is talented people that can build research harness robots in the physical world, PCB designers and firmware developers
@_cherki82_ they're training redditors, i'm training predators
When I say robotics, I mean control systems. I mean things like grasping items, moving, actuating to achieve some objective. I could probably solve it, I think. I would need more than I have, but really not that many GPUs
@cowboykaizen mujoco playground
@yacineMTB @jsuarez
The way I'm doing AI robotics is completely and totally different from all the super funded silicon valley companies. The reason I'm doing it differently is because they're all wrong
@yacineMTB Brother LLMs are still neural networks. U can train a small qwen to solve cartpole. U might be looking at this wrong. Lmk if u wanna chat
@mov_axbx @shakoistsLog i'm just qqing also i need physical access. i can't explain it
@yacineMTB I agree, the correct way to actually make robots that are alive.
@yacineMTB @agenticmachine @khoslaventures It’s because you don’t have anything actually built. You’re a larper farming X engagement
@yacineMTB the software/intelligence stack is where all the problems lie not in the fucking actuator/hardware stacks but that’s just my take 🤷♂️
@shakoistsLog shut up i need more capital for more gpus not even that much like 50 bones
it was solved but every roboticist has been nerd sniping themselves or the next one for like 40 years.. re-doing the stack again in a different medium.. pneumatic hands with great solenoid valve and even digital controls from the 60s were already there.. the guys probably hand built their own lathes back then, just sniped the whole way down for no reason other than research labs be labbing
@yacineMTB
The fact that robotics isn't unilaterally solved is just an algorithms gap. All of the silicon valley companies doing "robotics" right now with their data collection meme are on the wrong path entirely.
It's surprising that robotics isn't really, actually solved. It's not hard
@yacineMTB do u have an intro guide? i want to learn
@agenticmachine @khoslaventures i don't know if i want to raise. idk. maybe give me your contact or something idk though i'm not that sure about it like. maybe i just don't
@yacineMTB @khoslaventures fund him plz
@yacineMTB @shakoistsLog I offered you the use of a rack of 7x4090
@BenderDrummer @yacineMTB Oooofff... we still have a lot to figure out
@shakoistsLog @yacineMTB bro lives in constant delusion. Poor guy