I find the term “prompt engineering” kind of nauseating when it’s tech bros talking to me about how to use LLMs better (even though they have valid points). But within the context of Anthropic and...
I find the term “prompt engineering” kind of nauseating when it’s tech bros talking to me about how to use LLMs better (even though they have valid points). But within the context of Anthropic and other companies actually engineering their model system prompts it’s a very apt term. This is all very interesting and makes me understand Claude’s tone and behavior better; I didn’t know this was publicly available!
I find the term “prompt engineering” kind of nauseating when it’s tech bros talking to me about how to use LLMs better (even though they have valid points). But within the context of Anthropic and other companies actually engineering their model system prompts it’s a very apt term. This is all very interesting and makes me understand Claude’s tone and behavior better; I didn’t know this was publicly available!