Question 1

What is the primary goal of prompt engineering?

❌ To eliminate the need for model training
❌ To replace traditional programming techniques
❌ To minimize computational resources
✅ To design inputs that effectively guide the model to produce desired outputs

Explanation:
Prompt engineering focuses on crafting prompts that steer the model to give accurate, useful results.

---

Question 2

Correct prompt element and function match:

❌ Output indicator – wrong
❌ Instructions – wrong
✅ Input data – To provide the specific content the LLM needs to process
❌ Context – wrong

Explanation:
Input data refers to the exact text the model must analyze or transform.

---

Question 3

Primary function of format() in a PromptTemplate:

❌ Translate languages
❌ Convert to JSON
✅ Replace placeholder variables with actual values
❌ Optimize token count

Explanation:
format() fills template variables, creating the final prompt sent to the LLM.

---

Question 4

Which is NOT a step in creating an LCEL pattern?

❌ Defining a template
✅ Pre-processing all input data to ensure type compatibility
❌ Using pipe operator
❌ Creating a PromptTemplate

Explanation:
LCEL does not require pre-processing all data; it handles types internally.

---

Question 5

How does in-context learning help LLM adaptation?

❌ Requires weight updates
✅ Allows learning new tasks from a few examples without training
❌ Requires fine-tuning
❌ Limits LLMs

Explanation:
In-context learning means the model learns from examples given in the prompt—no training needed.

---

Question 6

How should Sarah improve AI customer service responses?

❌ Avoid context
✅ Use clear instructions + rich context
❌ Rely only on pre-trained knowledge
❌ Use vague prompts

Explanation:
Good prompts + context help the model understand the customer scenario and respond properly.

---

Question 7

Purpose of the pipe operator in LCEL:

❌ Convert functions
❌ Execute in reverse
✅ Connect components into a readable flow of data
❌ Replace RunnableParallel

Explanation:
The pipe operator (|) creates logical chains: Prompt → Model → OutputParser.

---

🧾 Summary Table