But a harness can also be: Claude Code, OpenClaw, Hermes, or Notion. The attached article helps readers understand the basic mechanism around tool_calling that the aforementioned AI harnesses are leveraging. But if you don’t know your own tools, you certainly won’t know the tools AI models leverage (for better or worse).
Thank you so much for your time btw I really appreciate your feedback, and I think what I’m trying to understand is one layer upstream. My brain keeps going to the ecology that produces the harness. I’ve been looking at a lot of adaptive systems through a cybernetic lens, not because I think biology has the answers, but because it gives me a common language for comparing systems that look completely different. I keep wondering whether having that shared language would make certain organizational problems easier to recognize before they become degradation. I honestly don’t know if people in AI are already thinking about it that way.
Many are! Cybernetics has of course been around for decades now (as a topic/subject). So you just need to find your way to those circles. Places like X, Reddit, ArchivX, LessWrong, are some good places to connect in with folks who’ve already been thinking about this (so you’re not working in isolation and find a real gap to address).
here?
Yes, this is one example.
But a harness can also be: Claude Code, OpenClaw, Hermes, or Notion. The attached article helps readers understand the basic mechanism around tool_calling that the aforementioned AI harnesses are leveraging. But if you don’t know your own tools, you certainly won’t know the tools AI models leverage (for better or worse).
Thank you so much for your time btw I really appreciate your feedback, and I think what I’m trying to understand is one layer upstream. My brain keeps going to the ecology that produces the harness. I’ve been looking at a lot of adaptive systems through a cybernetic lens, not because I think biology has the answers, but because it gives me a common language for comparing systems that look completely different. I keep wondering whether having that shared language would make certain organizational problems easier to recognize before they become degradation. I honestly don’t know if people in AI are already thinking about it that way.
Many are! Cybernetics has of course been around for decades now (as a topic/subject). So you just need to find your way to those circles. Places like X, Reddit, ArchivX, LessWrong, are some good places to connect in with folks who’ve already been thinking about this (so you’re not working in isolation and find a real gap to address).