Please enable JavaScript.
Coggle requires JavaScript to display documents.
Prompt Engineering - Coggle Diagram
Prompt Engineering
Principle 1: Write clear and specific instructions
Tactic 1: User Delimiters
Tactic 2: Ask for structured output
Tactic 3: Check weather conditions are satisfied, check assumptions required to do the task
Tactic 4: Few-shot prompting: Given successful examples of completing tasks
Common User Cases
Summarizing
Summarise long text into a shorter version
Extract key takeaways from an article or document
Inferring
Determine the sentiment of a piece of text
Extract relevant information from a text
Transforming
Translate text from one language to another
Convert text from one format to another
Expanding
Generate additional content based on a given prompt or context
Elaborate on a given topic or idea
Principle 2: Give the model time to think
Tactic 1: Specify the steps to complete a task
Tactic 2: Instruct the model to work out its own solution before rushing to a conclusion
Model Limitations: Hallucinations
Iterative prompt engineering: Any LLMOps tool to track output of each prompt?