Re:Malleable Software in the Age of LLMs (2023)
by XenophileJKO on Tuesday April 30, 2024 @03:33PM
(#40215262)
Tell me about it... I work on prompting a lot and you would be surprised how different the model interprets things like "suggestion" vs "recommendation".
Prompt engineering and getting structured data in and out of an LLM naturally lead you to something like precise English with JSON and pseudocode syntax sprinkled in