Mark this lesson as complete

Finish all lessons and pass the quiz to earn your official course completion certificate.

Instruction-based prompts

Video Module 4: Prompt structures & patterns

The Anatomy of a Great Prompt:- To get the best out of an AI, a high-quality instruction-based prompt usually contains four key pillars:

Component	Purpose	Example
Persona	Sets the tone and expertise level.	"Act as a senior data analyst..."
Task	The specific action to be taken.	"...summarize this spreadsheet into three trends."
Context	Background info or constraints.	"The audience is a board of directors who value brevity."
Format	How the output should look.	"Use a bulleted list and a concluding table."

Common Techniques

Few-Shot Prompting:- Instead of just asking for a result, you provide a few examples of what a "good" answer looks like.
- Prompt: "Classify the sentiment. Example: 'I love this!' -> Positive. 'It's okay.' -> Neutral. 'I hate this.' -> Negative. 'The service was slow.' ->"
- Result: The AI follows the pattern precisely.
Chain-of-Thought (CoT):- You ask the model to "think step-by-step." This is crucial for complex logic or math.
- Instruction: "Solve for $x$: $3x + 5 = 20$. Show your work step-by-step."
- Result: The AI breaks down the logic, reducing the chance of a "hallucinated" or incorrect final number.
Constraint-Based Prompting:- Setting boundaries to prevent "rambling" or off-topic content.
- Instruction: "Explain quantum entanglement without using the word 'spooky' and keep it under 50 words."