Post

APO relies on learning from feedback to optimize prompts, beginning with data and a simple, initial prompt. The data with this prompt is fed into the LLM model to generate responses, which are then analyzed by an evaluation model.
See the workflow → goo.gle/3NI9pxS

English



