# Best Practices Based on our experience at Prst.ai, we've identified several best practices to help you succeed in prompt engineering: ## **Evaluate Prompts with SMEs** Involving subject matter experts (SMEs) significantly reduces the risk of generating invalid or irrelevant outputs, thereby enhancing prompt quality and the business value of your product. While SMEs may not need to know the details of the prompt itself, presenting only the generation result during experiments can keep the process clear and focused. ## **Use Version Control** Usage of prompt version control enables you to store changes history and information about evaluations at each stage of the prompt creation lifecycle. Creating a new prompt version during each [Prompt Loop](/prompt-engineering/prompt-loop.md) iteration allows for easy comparison and understanding of which version performed better. > 💡 **Note:** Read More about Version Control in Prompts Section ## **Regular Re-Evaluation**: Even if you believe you have the best prompt that meets all your requirements, it's still important to regularly re-evaluate it. Testing prompts with a wide variety of input parameters is crucial to uncovering any potential issues or incorrect results. Without comprehensive testing, unpredictable outcomes may occur. --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.prst.ai/prompt-engineering/best-practices.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.