1. Zero-Shot Prompting Definition: Directly instruct the AI to perform a task with no examples. Why it works: Relies on the model’s general knowledge to understand your intent. Example Prompt: “Classify the following movie review as Positive, Negative, or Neutral. Review: ‘The storyline was engaging but the dialogue felt flat.’ Sentiment: ”Outcome: Correctly labels sentiment as Neutral, using just minimal guidance. 2. Few-Shot Prompting Definition: Include a few examples to demonstrate the format or style you expect. Why it works: Helps the model mimic structure and tone. Example Prompt: “Example: A 'baku' is a large blue flightless bird. ‘We saw many bakus on our trip.’ Now: ‘Write a short story about a baku that ended up on a spaceship bound for Mars." Uses the pattern to generate a coherent, stylistically consistent story. 3. Chain-of-Thought (CoT) Prompting Definition: Encourage step-by-step reasoning to solve complex tasks. Why it works: Breaks tasks into smaller logical parts, improving accuracy. Example Prompt: "I had 8 marbles, gave 3 away, then found 4 more. How many do I have now? Think step by step." Outcome: 1. You started with 8. 2. Gave away 3 → 5. 3. Found 4 more → 9.Answer: 9. 4. Meta Prompting Definition: Guide with abstract, structured steps instead of specifics—great for repeatable logic. Why it works: Keeps prompts flexible and reusable across tasks. Example Prompt: “Step 1: Define variables. Step 2: Apply formula. Step 3: Simplify and solve.” Outcome: Useful for generic coding problems or analytical processes. 5. Self-Consistency Prompting Definition: Generate multiple reasoning paths (CoT) and select the most consistent answer. Why it works: Combats reasoning errors and improves reliability. Outcome: Produces more accurate results by cross-checking across logic chains. 6. Context-First Prompting (Agent Focused) Definition: For AI agents, provide rich context to guide decision-making. Why it works: Agents perform better when they have clear, relevant background.