Final Exercise: Chain Prompting

Prompt Chaining

by PromptingGuide.ai

To improve the reliability and performance of LLMs, one of the important prompt engineering techniques is to break tasks into its subtasks. Once those subtasks have been identified, the LLM is prompted with a subtask and then its response is used as input to another prompt. This is what's referred to as prompt chaining, where a task is split into subtasks with the idea to create a chain of prompt operations.

Prompt chaining is useful to accomplish complex tasks which an LLM might struggle to address if prompted with a very detailed prompt. In prompt chaining, chain prompts perform transformations or additional processes on the generated responses before reaching a final desired state.

Besides achieving better performance, prompt chaining helps to boost the transparency of your LLM application, increases controllability, and reliability. This means that you can debug problems with model responses much more easily and analyze and improve performance in the different stages that need improvement.

✍ What You’ll Practice

In this final exercise, you’ll use a generative AI chat tool (like Copilot, ChatGPT, Gemini, or Claude) and bring together everything you’ve learned to create an original AI-assisted review of Great Expectations by Charles Dickens. You’ll use prompt chaining to guide the AI through a structured process and complete 7 steps that take you from identifying key themes to generating and formatting your final review. Each step is designed to help you practice writing effective prompts, evaluating AI responses, and refining your ideas.

🚩 Checkpoints:

For each step, there will be a checkpoint in the submission form asking for specific information. We recommend keeping the form open while you work through the exercise so you can easily track the questions and input your responses as you go.

💡 Tips

• Fact Check: Always verify the AI’s responses using trusted sources. If something seems off, research and revise.
• Engage Actively: Don’t hesitate to ask follow-up questions or try different angles. Review the Best practices and How to Prompt Introduction pages.
• Document Your Process: Keep a record of your prompts and the AI’s responses. This will help you reflect on your learning and improve your prompting skills.
• Use AI Suggestions: Many AI tools offer follow-up prompt ideas. Consider exploring them to deepen your understanding.

🛠️ Choosing Your AI Tool

For this exercise, you can use any generative AI chat (like Copilot, ChatGPT, Gemini, or Claude). The chosen tool needs to:

• Allow long conversations on the same chat (you’ll need to build on previous prompts),
• Support document export (PDF, Word, or text),

As SLC students, we recommend you to use Copilot through your school account. It’s integrated with your Microsoft package, supports document export, and won’t use up credits from other platforms. However feel free to choose which ever you feel more confident and comfortable with.

✅ What You Need to Do:

1. Summarize the Book's Themes
Ask Copilot to summarize "Great Expectations" into 4 main themes. Ensure the response provides enough detail to understand the book's general opinions on these themes.
🎯Goal: Learn how to craft effective prompts that elicit detailed and informative responses from AI.
💡Tip: Be interactive and ask for clarifications and direct quotes from the book so you have a better understanding of the topics. Ask for references and check them.

2. Deep Dive into a Theme
Choose the theme you find most interesting and ask Copilot for further explanations. Compare the book's perspective on this theme with other viewpoints you've encountered. Consider what new ideas it brings, any lacking points, and how it is portrayed in the book.
🎯Goal: Develop skills in asking follow-up questions and analyzing AI responses for deeper understanding.
💡Tip: Be critical in your answers and ask for further explanations on generic responses. Often, multiple iterations are needed to move beyond 'AI-like' answers. If a new concept is introduced, ask for clarifications, its relevance to the theme, and exactly how they appear in the book.

3. Compare with Another Media
Find another piece of media (book, article, movie, series) that also discusses the chosen theme. Ask Copilot to summarize and compare the two points of view. You can give AI a heads-up on what connection you see between the theme and the media. It helps generate more relevant comparisons.
🎯Goal: Practice using AI to draw comparisons and synthesize information from multiple sources.
💡Tip: Choose something meaningful to you. Ask the AI for suggestions if needed, but reflect on your own experiences too.

4. Reflect and Question
Reflect independently on the information gathered. Consider whether you agree with everything, if something feels wrong, or if anything sparks your curiosity. Ask Copilot about your doubts or request further explanations on interesting points.
🎯Goal: Strengthen your critical thinking and dialogue with AI.
💡Tip: Ask for examples from the book and the media you chose. Use this step to refine your understanding and assessing the veracity of the information discussed.

5. Create an analytical review
Ask Copilot to summarize "Great Expectations," incorporating previous information and comparing it to the other piece of media. Ensure your summary is unique by including your own insights and highlighting the parts you found most curious (and why).
🎯Goal: Learn how to refine prompts and use AI to create personalized and comprehensive summaries.
💡Tip: Don't expect to achieve the final summary in one prompt. Make modifications based on AI responses until you are satisfied. Sometimes, it is better to re-write and ask for modifications on top of your own summary.

6. Adapt the Language to a Specific Context
Choose a context (e.g., academic, professional, creative) and tailor the language of your final summary to fit this specific context and audience.
🎯Goal: Learn how to adapt your writing to different situations and audiences, enhancing the relevance and impact of your text.
💡Tip: Include a brief introduction that explains the context. For example, "As a college student taking a Literature course..." or "As a book critic for a renowned magazine...". Providing the AI with a clear context can help ensure the language and tone are appropriate for the intended audience.

7. Export Your Review
Ask AI to export your final review as a PDF, DOC (Word), or TXT file. The final document needs to be readable, appropriately looking, and be less than 2 pages long.
🎯Goal: Learn how to finalize and format your work for sharing.
💡Tip: Double-check the formatting and make sure the file opens correctly.

📩 Submission

Answer the form below by filling in the checkpoints for each step of the exercise and sharing your personal reflection on the process.

Learning Resources >>