Hi everyone, I've spent the last hours drafting a conceptual architecture for a new type of multi-agent AI network I call ASAN (Autopoietic Specialist-Agent Network).
It combines ideas like MoE, autopoiesis, QLoRA for 'rapid agent birth', and a governance layer. I've just put the full specification on GitHub (my first uplaode there) and would be incredibly grateful for feedback, criticism, and new ideas from this community. Thank you!
Variable-Fox•6h ago
It combines ideas like MoE, autopoiesis, QLoRA for 'rapid agent birth', and a governance layer. I've just put the full specification on GitHub (my first uplaode there) and would be incredibly grateful for feedback, criticism, and new ideas from this community. Thank you!