Prompt Engineering is for Transactional Prompting
Read OriginalMitchell Hashimoto clarifies the concept of prompt engineering by differentiating interactive prompting (like ChatGPT conversations) from transactional prompting (treating LLMs like programmable functions). He argues prompt engineering's core value lies in transactional use cases, where the goal is high-accuracy, low-cost outputs for objective tasks like classification or code generation, as opposed to subjective creative tasks.
Comments
No comments yet
Be the first to share your thoughts!
Browser Extension
Get instant access to AllDevBlogs from your browser