Useful Generative AI Prompt Techniques for Everyday Work

Written by Coursera • Updated on

Follow this practical guide to grow your AI skills, better integrate GenAI into your processes, and prepare for your future job role.

Jeff M - A graphic illustrating that GenAI can be used to envision a future job role.

Generative AI (GenAI) can be a powerful tool for enhancing productivity and creativity in everyday work tasks. In this guide, we’ll offer practical prompt techniques that you can use to leverage AI effectively, helping you with everything from summarizing documents to crafting emails.

Drawing on insights and key processes outlined in the course Use Generative AI as Your Thought Partner taught by Coursera CEO Jeff Maggioncalda, this guide is designed to be a valuable resource for professionals at all levels. Use this as a reference as you continue exploring large language model (LLM) tools and growing your GenAI skills.

Understanding prompt types

There are three basic prompting styles: basic query, context-enhanced, and thought-partner mode. Each style serves a different purpose and can be used depending on the complexity of the task.

Use basic query mode for quick, general inquiries. It’s ideal for straightforward questions that don’t require additional context. In context-enhanced mode, you provide specific context to the LLM for more accurate and relevant responses. Use thought-partner mode for complex problem-solving and strategic thinking. Start with your idea or point of view and ask the LLM to improve, challenge, or iterate on it.

In this guide, we detail Maggioncalda’s process for context-enhanced and thought-partner modes.

Placeholder

Step 1: Choose the right LLM.

In the course, you have access to Coursera’s secure GenAI Playground, which runs on Coursera’s enterprise access to Google Gemini Pro according to the data security and privacy policies of Google Cloud Platform services. Nothing you enter will be recorded, stored, or accessible by any Coursera employee at any time. And nothing you enter can or will be used by Google, Coursera, or any other entity to train any AI model.

Outside of the course, selecting the right LLM for your needs is essential for effectively using GenAI. Consider the specific needs of your work and the data security and privacy policies of each tool. If you’re planning to use GenAI at work, be sure to do so in compliance with your company’s technology policies.

Recommended GenAI tools:

  • Google Gemini: Offers advanced features and works smoothly with Google services, making it a powerful tool for those already using Google's ecosystem.

  • OpenAI’s ChatGPT: Known for its versatility and supported by a large community, making it accessible and adaptable for various applications.

  • Microsoft Copilot: Provides strong integration with Microsoft products, making it an excellent choice for users who rely on Microsoft software in their daily tasks.

  • Anthropic Claude: Known for its high safety and ethical standards, making it a good enterprise option for responsible and secure AI solutions.

Step 2: Input your context.

Your context is the background information that your LLM will need in order to adequately respond to your question. For example, if you want the LLM to optimize your resume for a specific job role, you may decide to provide context such as your current resume and your target role’s job description.

To input your context, use the prompt:

Consider the following CONTEXT and reply "I understand the context" but do not explain: 

<<< BEGIN CONTEXT >>>

[add context here]

<<< END CONTEXT >>>

Repeat this process for each piece of context you are providing.

Step 3: Input your question.

After adding your context, you’re ready to use your LLM as a thought partner. Here are some use cases and sample GenAI prompts for everyday work:

Summarizing documents

  • Summarize the key points of CONTEXT, focusing on key aspects and implications for [goal].

  • Explain the CONTEXT by breaking it down into simple terms.

  • Compare CONTEXT A and CONTEXT B, identifying key similarities and differences.

Drafting reports and communications

  • Use the CONTEXT to generate [output] about [topic] with the goal of [goal].

  • Compose a [type of communication] that most effectively communicates CONTEXT to [audience] in [style].

  • Incorporate key elements of CONTEXT A into CONTEXT B with the goal of [goal].

Analyzing data and information

  • Analyze the CONTEXT by examining its components and their relationships.

  • Evaluate the degree to which CONTEXT is [criteria] and assess its pros and cons in this regard.

  • Advise me on how to improve CONTEXT in order to better [goal].

Advanced prompting techniques

Using more advanced techniques can help you get even better results from your LLM.

Combining actions: Ask the LLM to perform multiple actions in a single prompt. Example prompt: Analyze the BCG article and advise on points that should be incorporated into our three-year strategy.

Iterative refinement: Continuously refine your prompts based on the feedback and responses you get.

Personalization: Customize the LLM's output to match your writing style or organizational tone. Example prompt: Consider the following writing style and reply, "I understand the style," but do not explain. <<< BEGIN STYLE >>> [insert writing sample] <<< END STYLE >>>

Placeholder

Step 4: Ensure accuracy and reliability.

Generative AI can recombine inputs in ways that may not be valuable or true. Here are five key actions to ensure the accuracy and reliability of AI outputs:

  1. Reflect: Think critically about each response. Does it make sense? Does it match your intuition?

  2. Validate: Confirm the factual accuracy of the information, especially for high-stakes decisions.

  3. Debate: Challenge the AI's responses and let it challenge your thinking.

  4. Filter: Request multiple options and sift through them to select the best ones.

  5. Integrate: Decide what to incorporate into your thinking, ensuring you remain accountable for the final decisions.

Keep growing your AI skills

By following these guidelines and techniques, you can effectively use GenAI as a powerful tool in your everyday work. Start practicing these skills now, and as GenAI continues to evolve, your ability to harness its strengths will only grow. After completing the introductory course Use Generative AI as Your Thought Partner, here are more courses to continue developing your AI skills.

Explore more GenAI basics with Google AI Essentials. In this nine-hour course, you’ll learn how LLMs work, how to use AI tools effectively, and how to stay up-to-date on the latest AI technologies.

Practice more prompting techniques with Vanderbilt University’s Prompt Engineering Specialization. Over three courses, you’ll explore more complex prompting techniques and how to generate more trustworthy responses from your LLM.

Expand your AI knowledge with IBM’s AI Foundations for Everyone Specialization. This beginner-friendly series builds upon fundamental AI skills and culminates in building your own chatbot, no coding experience required.

Eager to try all three? Consider a Coursera Plus subscription. With Coursera Plus, you’ll gain access to over 7,000 courses, Specializations, and Professional Certificates from top universities and industry leaders—all for one monthly or annual fee.

Coursera Plus
Build job-ready skills with a Coursera Plus subscription
  • Get access to 7,000+ learning programs from world-class universities and companies, including Google, Yale, Salesforce, and more
  • Try different courses and find your best fit at no additional cost
  • Earn certificates for learning programs you complete
  • A subscription price of $59/month, cancel anytime

Updated on
Written by:

Coursera

Writer

Coursera is the global online learning platform that offers anyone, anywhere access to online course...

This content has been made available for informational purposes only. Learners are advised to conduct additional research to ensure that courses and other credentials pursued meet their personal, professional, and financial goals.