In today’s world, it is crucial for students to engage with technology not just as users, but as creators. This assignment is designed to bridge this gap, offering students a hands-on opportunity to delve into the world of AI and chatbot development. The goal is to go beyond just programming or prompting a bot by emphasizing the importance of understanding audience needs, harnessing the power of structured prompting, and respecting ethical considerations. By building their own chatbots, students will not only develop technical skills but also enhance their critical thinking, problem-solving, and creative abilities.
➡️ Here is a free Poe tutorial to help you or your students started.
In the world of AI and chatbots, the key to success lies in precision and specificity. For this assignment, your challenge is to create a chatbot that is finely tuned to a super-narrow audience and use case. The goal is not just to use AI, but to master it by constraining it to a specific field of knowledge through structured prompting and tailored data. Your task is to sculpt AI's broad capabilities into a tool that consistently produces useful and relevant responses for your unique Substack audience.
Throughout this semester, you've cultivated a writing habit, engaged with your audience, crafted detailed reader personas, and gained a clear understanding of who might read and subscribe to your Substack. These efforts have laid the foundation for a robust chatbot. The insights you've gathered are invaluable for this endeavor. You're now equipped with the precise knowledge needed to create a chatbot that doesn't just respond, but resonates with your specific audience.
Assignment
Using Poe AI, design and build a chatbot that performs a valuable task for your Substack readers. This task should be based on the audience persona you've developed and the content themes you've explored so far.
You will need to take a structured approach to your chatbot. You can read more here, but this is the process in a nutshell.
Identify the Task: Choose a specific, useful task for your chatbot. This could range from answering FAQs about your content, offering personalized reading suggestions, to engaging in basic discussions related to your blog's theme.
Structured Prompting: Use a structured approach to build your chatbot. Define clear roles, goals, constraints, and examples in your prompts to guide the chatbot’s responses.
Development and Testing: Build your chatbot using an AI platform, and thoroughly test it. Make sure the chatbot accurately and effectively performs the chosen task.
User Experience Focus: Ensure your chatbot clearly communicates its purpose to the users and is easy to interact with. Pay attention to user feedback during testing to refine the chatbot.
Documentation: Write a brief report explaining your chatbot’s purpose, the task it performs, and how you used structured prompting in its development. Include insights from the testing phase and any revisions you made.
Examples
Choosing the right task for your chatbot is crucial. It should be a task that aligns with the interests and needs of your Substack audience, based on your semester-long engagement and understanding of their preferences. Here are some good examples from our online discussion.
Mental Wellness Assistant: A chatbot designed for those seeking a healthy mindset and mental stability, offering scheduling tips, fun activities, and at-home remedies to alleviate stress and anxiety.
Reading Material Recommender: A chatbot that recommends reading material based on criteria such as topic, genre, era, and user's previous reading material, suitable for both fiction and non-fiction readers.
Legal Precedent Researcher: A bot aimed at law students or aspiring attorneys, providing detailed information on major and minor court cases, helping with research and understanding of legal precedents.
Fitness Guide for Beginners: A chatbot catering to college women new to fitness, offering helpful tips, workout ideas, and encouragement for those hesitant to start their fitness journey.
Financial Planning Assistant: A chatbot designed for individuals seeking financial advice, capable of helping set up budgets, researching investment choices, and planning for future financial goals.
Ethical Considerations and Data Privacy
When developing your chatbot, it's crucial to consider ethical implications and data privacy. Remember, any data inputted into or handled by your chatbot is not inherently secure. As creators, you must respect the privacy and confidentiality of user information. Avoid collecting sensitive personal data, and be transparent with your audience about the data your chatbot collects and how it is used. When using public content for your knowledge base, that content should be cited either in the output or in the chatbot description.
Report Guidelines
Your documentation report should succinctly cover the key aspects of your chatbot project and follow memo report guidelines:
Heading: Include 'To:', 'From:', 'Date:', and 'Subject:' lines.
Introduction: Briefly state the purpose, objective, and target audience of your chatbot.
Chatbot Design and Development: Summarize the process of choosing the task, implementing structured prompting, and technical details.
Testing and Refinement: Outline your testing approach, feedback incorporation, and any adjustments made.
User Experience: Describe the chatbot's interface and user interaction.
Ethical Considerations and Data Privacy: Mention how you addressed ethical use and data privacy.
Conclusion and Learning Outcomes: Conclude with a brief overview of achievements, challenges, and key learnings.
This report should provide a concise overview of your chatbot development process, emphasizing key decisions, results, and insights.
Specifications
To be marked complete, this chatbot should:
Perform a clearly defined task relevant to your Substack content.
Use structured prompting effectively.
Include a user-friendly interface and clear instructions for users.
Be accompanied by a report documenting the development and testing process.
Submit your chatbot along with the documentation report for peer review. After incorporating feedback, you will present your chatbot and its functionalities to the class.
To get your bot builder badge, submit your documentation report to canvas with a share link to your bot.
I like a good challenge. The "Reading Materials Recommender" seems particularly enticing🤔