Best practices for writing AI prompts
15 min
Beginner
By the end of this lesson, you’ll be able to:
- Recognize the key components of a great prompt
- Write effective AI prompts
- Use follow-up prompts to refine AI results
The basics of AI prompting
What is AI?
Artificial intelligence (AI), refers to systems that can perform tasks that typically require human intelligence. AI can understand natural language, recognize patterns, and make intelligent decisions. Generative AI and AI chat tools can assist with a wide range of tasks to improve and streamline your workflow.
Generative AI can help you:
- Draft new documents and communication
- Summarize, edit, and improve content
- Generate new ideas and insights
- Organize, synthesize, and analyze data
- Make recommendations and suggestions
- Research unfamiliar topics
What is a prompt?
A prompt is a piece of text or instruction given to an AI to generate a response. Think of it as a way to communicate with the AI, guiding it to produce your desired output. Prompts can range from simple questions to highly detailed instructions or even code. Understanding how to craft effective prompts is crucial because the quality of the AI's response is directly tied to the quality of the prompt.
Prompt engineering is the iterative process of designing and refining prompts to achieve the best possible results from an AI. It involves understanding the components of a good prompt and how to structure them effectively. This skill is essential for anyone looking to use AI tools for various tasks, from content creation to data analysis, and, like any skill, it takes practice to master.
Write a great AI prompt
The elements and level of detail included in your prompt should be tailored to the complexity of the task you want the AI to perform. If you are asking the AI a simple question, you can do that using the same natural language that you would use if you were asking a colleague. While AI is great for answering these types of questions, there is much more power that you can unlock by incorporating AI into your more complex workflows.
A well-crafted generative AI prompt follows TCREI framework, which stands for:
- Task: Define the outcome you are expecting from Rovo.
- Context: Provide helpful background information that can help Rovo understand the scenario.
- References: Include any data (e.g. attachments) or links that are relevant to guide Rovo in generating its response.
- Evaluate: Assess Rovo’s response for accuracy and relevance. Identify if something is missing or is unclear.
- Iterate: Use Rovo’s response and your evaluation to refine your prompt. This is where you add more detail, clarify your request or break it down into simpler requests.
Using this technique helps create accurate and effective prompts.
Let’s look at an example!
Bad prompt:
"Why is my Wi-Fi slow?"
Good prompt that add context:
"My Wi-Fi is slow at home. Can you help me fix it?"
Great prompt that uses TCREI:
"You are a technical support agent. My home Wi-Fi is slow and keeps dropping connection, especially in the evenings. I use a Netgear Nighthawk router and live in a two-story house. Please provide a step-by-step troubleshooting guide in bullet points, prioritizing the most common causes first. If you need more details, let me know what to check."
Here is why this is a great prompt:
Here is why this is a great prompt:
- The Task is properly defined: Asks for a step-by-step troubleshooting guide.
- Context is provided by describing the environment.
- The Reference here is offering to provide more detail if needed.
- Evaluate and Iterate: The user can review the steps and provide feedback or additional info for further refinement.
In addition to the required components, there are several optional elements that can enhance the quality of your AI prompts and improve the AI’s response. These elements are not necessary for every prompt but can be useful in specific scenarios.
Depending on the type and complexity of the task you’re asking the AI to perform, a good prompt may include the following additional elements:
- Format: Specify the desired format of the output.
- Tone: Define the writing style or mood of the AI’s response.
- Examples: Provide good examples to guide the AI.
👇 Click the boxes below to explore each optional element in more detail and see examples.
Prompting best practices
When crafting prompts for AI, following best practices can significantly improve the quality and relevance of the responses.
👇 Click the tabs below to explore the dos and donts when crafting prompts.
✔️ Be specific and clear
✔️ Be concise and include relevant context
✔️ Use simple, natural language
✔️ Break highly complex tasks into a series of smaller prompts
✔️ Verify or test the AI results to ensure validity
✔️ Use follow-up prompts to iterate and refine the results
While AI can enhance productivity and streamline workflows, it's important to be aware of its limitations. First, AI can sometimes have hallucinations, or produce information that appears correct, but is in fact inaccurate or misleading.
Second, AI results are only as good as its knowledge base.
👉 For example: If you are using Rovo but don’t have sufficient third-party app connectors set up, or you don’t have permission to access the data, the quality of the output will be impacted. It is critical to always manually validate AI-generated results.
Continue the conversation with follow-up prompts
Not in love with the results you got from AI? Refine the results using follow-up prompts rather than starting over with a new prompt! Remember, prompting is an iterative process.
After receiving an initial response, you can provide additional prompts to clarify, expand, or adjust the output until the results meet your expectations. You can ask additional questions or provide further instructions based on the initial output. This conversational process helps achieve more accurate, relevant, and useful results.
Some tips for effective follow-up prompting:
- Be specific: Clearly state what needs to be changed or improved, such as the tone, format, or level of detail.
- Provide positive and negative critique: In addition to specifying what you’d like to see changed, mention what aspects of the initial output were good to ensure those elements are retained in the next iteration.
- Provide additional context or instruction: Offer more background information, context, or instructions to guide the AI in generating a more accurate response.
- Iterate as needed: Continue the process of reviewing and prompting until you achieve the desired result.
👇 Click on the tabs below to see examples of how follow-up prompts can be used to refine AI outputs.
Resolve ambiguities or unclear aspects of the initial response.
👉 For example: "Can you provide more details on the current risks and how they are being mitigated?"
How was this lesson?
next lesson
Best practices for using AI in Atlassian products
- Atlassian Intelligence results are based on existing data
- Searching work items in Jira using Atlassian Intelligence
- Using Atlassian Intelligence in Jira Service Management
- Give feedback