Session 2: The Art of Prompt Engineering for ESL Teachers
Welcome Back, Teachers!
In our previous session, we explored the fundamentals of Artificial Intelligence, Generative AI, and ChatGPT. We learned that AI is not a replacement for teachers but a powerful tool that can help us save time, generate ideas, and improve classroom preparation.
Today, we will discuss one of the most important skills in AI-assisted teaching: Prompt Engineering.
Many teachers start using ChatGPT and become disappointed with the results. They might ask the AI to create a lesson plan and receive a generic response. They might request a worksheet and find it unsuitable for their learners. As a result, they conclude that AI is not useful.
However, in most cases, the problem is not the AI.
The problem is the prompt.
Just as students need clear instructions to complete a task successfully, AI also needs clear instructions to produce high-quality outputs.
By the end of this session, you will learn:
- What a prompt is
- Why some prompts fail
- The anatomy of an effective prompt
- Role Prompting
- Context Prompting
- How to write professional prompts for ESL and IELTS teaching
Let us begin.
What Is a Prompt?
A prompt is the instruction, question, or command that you give to an AI tool.
Simply put, a prompt is how you communicate with AI.
Consider the following classroom situation.
A teacher says:
“Do the exercise.”
Students may ask:
- Which exercise?
- How many questions?
- Individually or in pairs?
- How much time do we have?
The instruction is too vague.
Now imagine the teacher says:
“Work in pairs. Complete Questions 1–10 on page 25. You have 10 minutes. Focus on using the Present Perfect Tense correctly.”
The task becomes much clearer.
The same principle applies when communicating with AI.
The quality of your output largely depends on the quality of your prompt.
In the world of AI, there is a common saying:
“Garbage in, garbage out.”
If your prompt is weak, your result will often be weak.
If your prompt is detailed and purposeful, your result is usually much better.
Why Some Prompts Fail
Let us examine a common mistake.
Bad Prompt
“Teach vocabulary.”
This prompt is extremely broad.
The AI does not know:
- Which vocabulary?
- Which level?
- Which age group?
- Which teaching method?
- How much time is available?
As a result, the response may be generic and not useful for your classroom.
Now compare it with this:
Better Prompt
“Act as an IELTS instructor. Create a 20-minute vocabulary lesson on environmental issues for B1-level learners. Include a warm-up activity, vocabulary presentation, practice task, and homework assignment.”
Notice how much information the AI receives.
The output becomes significantly more focused and practical.
Most unsuccessful AI interactions occur because users provide too little information.
Think of AI as a new teaching assistant.
If you give your teaching assistant vague instructions, you cannot expect excellent results.
The Anatomy of a Good Prompt
Professional prompts usually contain several key elements.
1. Role
Tell the AI who it should be.
Examples:
- Act as an IELTS instructor.
- Act as an experienced TESOL trainer.
- Act as a curriculum designer.
- Act as a primary school English teacher.
The role helps the AI adopt the appropriate perspective.
2. Task
Clearly state what you want.
Examples:
- Create a lesson plan.
- Generate a reading passage.
- Design a speaking activity.
- Write a vocabulary worksheet.
Avoid vague requests.
Be specific.
3. Audience
Who are the learners?
Examples:
- A1 learners
- B1 learners
- IELTS candidates
- Young learners
- University students
The AI needs this information to adjust complexity.
4. Topic
Specify the content area.
Examples:
- Climate change
- Business communication
- Travel and tourism
- Technology
- Health and fitness
5. Format
Tell AI how to organize the output.
Examples:
- Table format
- Lesson plan format
- Bullet points
- Worksheet format
- PowerPoint outline
6. Constraints
Provide limits.
Examples:
- 20-minute lesson
- 150-word reading text
- Ten vocabulary items
- Five discussion questions
These details improve accuracy.
The Formula for Strong Prompts
A simple formula is:
Role + Task + Audience + Topic + Format + Constraints
For example:
“Act as an experienced IELTS instructor. Create a 30-minute vocabulary lesson for B1-level learners on environmental issues. Present the lesson in a table format with learning objectives, activities, timing, and homework.”
This prompt is far more powerful than simply writing:
“Teach vocabulary.”
Understanding Role Prompting
Role Prompting is one of the most effective prompt engineering techniques.
In Role Prompting, you assign a professional identity to the AI.
Instead of asking:
“Explain the Present Perfect.”
You write:
“Act as an experienced ESL teacher and explain the Present Perfect Tense to A2-level learners.”
The difference may seem small, but the results are often much better.
Why?
Because the AI now understands the perspective from which it should respond.
Examples of Role Prompting for ESL Teachers
Example 1
“Act as a CELTA-certified ESL teacher. Create a communicative activity to teach phrasal verbs.”
Example 2
“Act as an IELTS examiner. Generate ten Speaking Part 2 topics.”
Example 3
“Act as a TESOL trainer. Explain Task-Based Language Teaching to novice teachers.”
Example 4
“Act as an English teacher in Bangladesh. Create a lesson suitable for HSC students.”
The more relevant the role, the more useful the response often becomes.
Understanding Context Prompting
Context Prompting means providing background information.
Imagine a doctor trying to diagnose a patient.
Without context, diagnosis becomes difficult.
Similarly, AI performs better when it understands the situation.
Consider these two prompts.
Weak Prompt
“Create a speaking activity.”
Strong Context Prompt
“I teach a class of 20 B1-level university students in Bangladesh. They are preparing for IELTS Speaking. The lesson duration is 30 minutes. Create a pair-work speaking activity on environmental issues.”
The second prompt contains context.
The AI can now generate a more realistic and relevant activity.
Why Context Matters in ESL Teaching
Every classroom is different.
Consider these variables:
- Age
- Proficiency level
- Learning objectives
- Cultural context
- Class size
- Lesson duration
When these details are included in prompts, AI becomes significantly more useful.
As professional educators, we should train ourselves to provide context every time we interact with AI.
Practical Activity
Now let us practice.
Prompt 1
“Teach vocabulary.”
Evaluate the response.
Would you use it directly?
Probably not.
Prompt 2
“Act as an IELTS instructor. Create a 20-minute vocabulary lesson on environmental issues for B1-level learners. Include a warm-up, vocabulary presentation, controlled practice, communicative activity, and homework.”
Compare the two outputs.
Notice how the second prompt produces a lesson that is more structured, relevant, and classroom-ready.
This simple exercise demonstrates the power of prompt engineering.
Reflection Questions
- How often do you give AI vague instructions?
- Which element of prompt writing do you find most challenging?
- How can Role Prompting improve your lesson planning?
- How can Context Prompting help create more suitable materials for your learners?
- What teaching tasks could benefit most from effective prompting?
Write your reflections in a teaching journal.
Key Takeaways
Today, we learned that prompt engineering is one of the most important skills for teachers using AI.
Remember these principles:
✓ A prompt is an instruction given to AI.
✓ Poor prompts often produce poor outputs.
✓ Effective prompts include role, task, audience, topic, format, and constraints.
✓ Role Prompting helps AI adopt the correct professional perspective.
✓ Context Prompting provides background information that improves accuracy.
✓ Better prompts lead to better teaching materials.
As ESL professionals, we should not aim to become AI experts. Instead, we should become expert communicators with AI. The teachers who master prompting will be able to generate lesson plans, assessments, worksheets, speaking activities, and classroom resources faster and more effectively than ever before.
In Session 3, we will learn how to use AI to create complete ESL lesson plans based on different teaching methodologies, including PPP, CLT, ESA, and Task-Based Language Teaching.
– Md Shakhawat Hossain
CEO & Founder, TalentHut TESOL Academy
LinkedIn Profile: https://www.linkedin.com/in/mdshakhawathossain/
#AIinEducation #PromptEngineering #ChatGPT #ESL #EFL #TESOL #ELT #EnglishTeacher #IELTS #TeacherTraining #TeacherDevelopment #EducationalTechnology #DigitalLearning #FutureOfEducation #LanguageTeaching #InstructionalDesign #AIForTeachers #ProfessionalDevelopment
