As a proof of concept I've been working on a Jev style vision model that can rapidly answer boolean and choice questions based on images (See [0] but not the point of this comment). One of the reasons I wanted to make this concept is because I was also thinking about exactly what the parent comment suggests. I just haven't gotten around to getting (3d printing) a robot arm yet :)
Mystery-Machine · · focus · HN ↗
thomasikzelf · · focus · HN ↗
WildGreenLeave · · focus · HN ↗
[0]: <a href="https://huggingface.co/MeerDevelopment/Qevi-2B" rel="nofollow">https://huggingface.co/MeerDevelopment/Qevi-2B
systemerror · · focus · HN ↗