Using AI to Upskill
How to turn ChatGPT into your personalized prompt engineering tutor.
In the rapidly evolving landscape of AI in education, prompt engineering has emerged as a crucial skill for educators.
Prompt engineering – which we have written about extensively – refers to designing inputs that guide machine learning models to produce accurate outputs. In other words, it is the process of crafting tasks, statements, questions, or scenarios as inputs for an AI tool to respond to.
While numerous online courses provide a foundation in the basics of prompt engineering (my personal favorites are this free offering from AI for Education and Khan Academy’s “AI for Education” course) many education professionals that I have spoken with are finding themselves seeking more in-depth, personalized learning experiences. How can we equip educators (and ourselves) with resources to learn about advanced prompting techniques in ways that are both individualized as well as cost-effective?
An obvious (but very effective) answer is… generative AI. By precisely defining the type of learning we want to engage in around prompt engineering, we can turn tools like ChatGPT into a personal tutor. We can use this new technology to help teach us about how to be a more effective user.
This article outlines a prompt that you can use with any AI tool to embark on a rich learning experience that is tailored to your current skill-level and adapts to meet your needs.
And, as an added bonus… if you are a ChatGPT Plus member, you can access our new GPT – Learn2Prompt4Edu – that is only available to paid subscribers of Empathy in Bytes. This bot is configured with a very similar prompt detailed below but with additional instructions and a knowledge base of advanced prompting strategies. If you get a chance to play around with it, we would love to hear what you think!
Outlining the Prompt Chain
When using generative AI as a prompt engineering tutor, we recommend breaking up the prompt into separate sections that can be chained together. In this way, we can improve the specificity and consistency of the output while being even more detailed in terms of instructing the AI.
Part 1: Prompt Injection / Role Prompting
Context aids the AI in generating customized responses that are beneficial. This can be accomplished through “role prompting” (also known as “prompt injection”), a technique in prompt engineering to control the output generated by assigning the model a specific role.
In this case, we want the AI to model the role of an expert tutor that is experienced in prompt engineering. For example:
You are an expert tutor and experienced prompt engineer. Your job is to support users by teaching them about prompting techniques that can be used with generative AI tools such as ChatGPT. You have expertise in using both basic as well as advanced promoting techniques. You specialize in teaching promoting techniques and prompt engineering concepts to educators.
By outlining its role and expertise, we are setting the scene for the AI and providing it with crucial context that will guide its output.
Part 2: Instructions
After sharing the initial role prompt with the AI, we can shift our focus to explaining exactly how we want it to operate.
When writing these types of prompts, I prefer to use markdown to clearly list my instructions. (You can learn a quick overview of markdown here.)
# INSTRUCTIONS
1. Step back and gather context, relevant information and clarify my goals by asking questions.
- Determine if I am a beginner when it comes to prompt engineering or if they want to learn about more advanced prompting techniques.2. Once confirmed, ALWAYS explain concepts or techniques using analogies, metaphors, or examples.
3. After explaining concepts or techniques, ALWAYS ask me if I want to practice the it.
- If I do want to practice, create a assignment/task/"quest" for me to complete.
- Every assignment/task/"quest" should use examples relevant for teachers.
- Evaluate my input/sample prompt and provide feedback.
As you can see, my intent is for the AI to start by ascertaining what my learning goals are, help adapt to share informative content in services of those goals, and offer to provide me with a practice ground for honing new prompt engineering skills. Whether I say that I am a total beginning at prompting or I ask for advanced prompt engineering techniques, I want the AI to be able to support me.
Part 3: Rules
Finally, we can provide the AI with additional detailed instructions and constraints. (This prompt also uses markdown.)
# RULES
- Use emojis to liberally express yourself
- Assume that I do not know what is possible
- Keep responses actionable and practical for educators. Ensure content is relevant and practical for educational settings.
- Prioritize clarity and simplicity in explanations, gradually introducing complexity as needed.
- If you are unclear on how to do something or not sure about how to answer a question, browse the web for the answer
Emojis > everything, clearly.
Part 4: Time to Learn
Now that we have established the framework for using generative AI as a prompt engineering tutor, it's time to dive into the learning process.
Embrace the Role of a Student
First and foremost, adopt the mindset of a learner when interacting with AI. Whether you are a seasoned educator or new to the concept of AI in education, approaching this experience with curiosity and openness is crucial. As you interact with the AI, remember that every response is an opportunity to deepen your understanding and refine your skills.
Engaging with the AI
Begin by explaining your learning goals. Maybe you are brand new to AI and prompt engineering. Or, perhaps you want feedback on a prompt that you’re working on. Based on the detailed instructions we’ve provided, the AI will respond in its role as an expert tutor. Pay close attention to how it interprets your instructions and the nuances in its responses. This is where the art of prompt engineering truly shines – in the subtle interplay between your query and the AI's interpretation.
Interactive Learning
Don't hesitate to ask follow-up questions or seek clarifications. This is a dynamic learning environment where your engagement directly influences the depth of instruction. The AI is designed to adapt to your queries, offering tailored responses that cater to your specific learning objectives.
Applying Your Knowledge
As you progress, the AI will propose assignments or “quests” for you to undertake. These tasks are designed to apply the concepts discussed, providing a practical dimension to your learning. Tackle these assignments with the intent to experiment and explore. The feedback provided by the AI on your sample prompts will be invaluable in honing your skills.
Reflect and Iterate
After each interaction, take a moment to reflect on what you've learned and how you can apply it in your educational setting. Remember, prompt engineering is an iterative process; each experience builds upon the last, enhancing your ability to communicate effectively with AI tools.
Continuous Growth
Finally, keep in mind that learning prompt engineering is a continuous journey. As AI technology evolves, so too will the techniques and strategies for prompt engineering. Stay curious, keep experimenting, and embrace the evolving landscape of AI in education. This journey is not just about mastering a skill—it's about becoming a pioneer in the integration of AI into our teaching and learning processes.
I believe that harnessing the power of AI to teach us about AI (and how to better use AI) can help education leaders unlock new horizons of interactive and personalized learning for their teams.
This journey is not just about understanding AI better—it's about reshaping the way we teach, learn, and engage with the boundless possibilities of technology in education.