Prompt engineering basics: Crafting effective prompts - Gemini Tutorial

- [Instructor] Effective prompting is a crucial part of getting the most out of Gemini. If you've ever received a weird or generic reply from a chatbot, it might just need a better question. And that's where prompting comes in. By the end of this lesson, you'll have the skills to be able to craft effective prompts that get useful and consistent results from Gemini. Let's look at what elements make an effective prompt. These are: being clear and concise, using a persona, for example, act as a copywriter to do X, Y, Z, providing context and background. It's very important to give the details surrounding the request. The task, it can be a single task or a series of tasks, like steps to follow. And lastly, refine. Correct and guide Gemini based on the responses that you receive. Gemini learns best by examples. Providing examples showing the content, style, structure, tone, or any elements you feel is important allows you to get the most out of Gemini and help it to understand how to give you what you want. Here's a structured way to craft prompts. It's what I call the BRAIN framework. To make it easy to remember, think of instructing the AI brain on what you want to do. B is for background. So this is where you set the scene and provide background information and context for Gemini to understand the request better. This is also called priming. An example would be, organize my day for maximum focus and efficiency. Next is R, which is for role assignment. And that defines Gemini's role. This is the persona to use for a more tailored response. The example: Act as a time management coach. Next, A is for action, which is to provide the action or task Gemini must complete, and the format can also be specified. So this is, put it in a bullet list, a numbered list, or a table, et cetera. And an example of that is, break my day into time blocks for work, exercise, and relaxation. Next is I, which is for inputs. And this is where you include the necessary details. This can also be specific requirements. For example, if you ask Gemini to write an article for you, the inputs would be things like writing style, important headings, number of words, et cetera. So in this case, we are specifying that it is for emails, planning, and a workout must be included. N is for narrowing down. This is where you refine the request and define the boundaries. This can also be for providing things like limitations. For example, prioritize deep work in the morning, but only 60 minutes, then relaxation in the evening. Then, putting it all together, you're able to see all the elements formed into a single prompt, as well as the output from Gemini, which has created a new schedule from all of our parameters. From this example and framework, you can see that clear, structured prompts mean better results. Use the BRAIN framework to prompt like a pro. So why not try it for yourself for your next Gemini prompt and see how the response is more structured and detailed.