, STUVIA EXAM LIBRARY
EXAMINATION DOCUMENT 2026/2027
WGU D685 OA Final Exam
Practical Applications of Prompt Engineering
Objective Assessment Complete Questions & Detailed Verified Answers
Original Study Guide - Different Look & Different Concept Per Question - Other Format Reused
,Question 1: [D685OA-Q001] PROMPT ENGINEERING CASE: prompt design patterns at Prompt
Engineering Center - Dr Jonah AI notes: Prompt: Few-shot with chain-of-thought. Focus: Prompt
engineering definition objective subjective assessment WGU D685. Choose best D685 answer.
A. Distractor: Violates Prompt engineering definition objective subjective assessment WGU D685 - misapplies
prompting or ignores safety for prompt design patterns
B. Prompt engineering definition objective subjective assessment WGU D685 - Correct: prompt engineering
definition objective subjective assessment wgu d685 applied to prompt design patterns per D685
C. Distractor: Confuses Prompt engineering definition objective subjective assessment WGU D685 with similar
LLM concept, misreads Prompt: Few-shot with chain-of-thought
D. Distractor: Claims Prompt engineering definition objective subjective assessment WGU D685 not needed for
prompt engineering prompt design patterns
Correct answer: B
Detailed Verified Rationale (Prompt engineering definition objective subjective assessment WGU
D685): Prompt case prompt design patterns at Prompt Engineering Center, Dr Jonah AI documents
Prompt: Few-shot with chain-of-thought. Prompt engineering definition objective subjective
assessment WGU D685 per WGU D685 OA Final Exam 2026/2027 Practical Applications of Prompt
Engineering Objective Assessment covers prompt engineering definition LLM fundamentals
transformer tokenization, prompt components instruction context, zero one few shot role chain of
thought zero shot CoT self consistency tree thought ReAct chaining automatic meta, parameters
temperature top p top k presence frequency max tokens stop sequences system user assistant
context window overflow injection jailbreak PII bias toxicity hallucination grounding RAG vector DB
embedding Pinecone chunking reranking evaluation BLEU ROUGE BERTScore perplexity human
LLM as judge cost latency optimization compression caching fine tuning LoRA business application
customer support content code summarization classification extraction translation Q&A; safety RLHF
filtering ethical data leakage adversarial versioning logging monitoring observ...
Question 2: [D685OA-Q002] Select BEST answer: Large language model LLM fundamentals
transformer architecture - AI Solutions Austin - LLM evaluation. Dr Priya LLM - Model: GPT-4
temperature 0.7 top-p 0.9.
A. Best: Large language model LLM fundamentals transformer architecture - Correct complete verified answer
per WGU D685 OA Final
B. Distractor: Violates Large language model LLM fundamentals transformer architecture - misapplies prompting
or ignores safety for LLM evaluation
C. Distractor: Confuses Large language model LLM fundamentals transformer architecture with similar LLM
concept, misreads Model: GPT-4 temperature 0.7 top-p 0.9
D. Distractor: Claims Large language model LLM fundamentals transformer architecture not needed for prompt
engineering LLM evaluation
Correct answer: A
, Detailed Verified Rationale (Large language model LLM fundamentals transformer architecture):
Prompt case LLM evaluation at AI Solutions Austin, Dr Priya LLM documents Model: GPT-4
temperature 0.7 top-p 0.9. Large language model LLM fundamentals transformer architecture per
WGU D685 OA Final Exam 2026/2027 Practical Applications of Prompt Engineering Objective
Assessment covers prompt engineering definition LLM fundamentals transformer tokenization, prompt
components instruction context, zero one few shot role chain of thought zero shot CoT self
consistency tree thought ReAct chaining automatic meta, parameters temperature top p top k
presence frequency max tokens stop sequences system user assistant context window overflow
injection jailbreak PII bias toxicity hallucination grounding RAG vector DB embedding Pinecone
chunking reranking evaluation BLEU ROUGE BERTScore perplexity human LLM as judge cost
latency optimization compression caching fine tuning LoRA business application customer support
content code summarization classification extraction translation Q&A; safety RLHF filtering ethical data
leakage adversarial versioning logging monitoring observability cost rate latency cac...
Question 3: [D685OA-Q003] Which is TRUE about Tokenization tokens token count cost
optimization? Prompt case at LLM Research Dallas, RAG, Dr Carlos data Evaluation: BLEU 0.78
ROUGE-L 0.82. WGU D685 OA Final.
A. Distractor: Violates Tokenization tokens token count cost optimization - misapplies prompting or ignores
safety for RAG
B. Distractor: Confuses Tokenization tokens token count cost optimization with similar LLM concept, misreads
Evaluation: BLEU 0.78 ROUGE-L 0.82
C. Correct for Tokenization tokens token count cost optimization: detailed verified per WGU D685 OA
2026/2027 practical prompt engineering
D. Distractor: Claims Tokenization tokens token count cost optimization not needed for prompt engineering RAG
Correct answer: C
Detailed Verified Rationale (Tokenization tokens token count cost optimization): Prompt case RAG at
LLM Research Dallas, Dr Carlos documents Evaluation: BLEU 0.78 ROUGE-L 0.82. Tokenization
tokens token count cost optimization per WGU D685 OA Final Exam 2026/2027 Practical Applications
of Prompt Engineering Objective Assessment covers prompt engineering definition LLM fundamentals
transformer tokenization, prompt components instruction context, zero one few shot role chain of
thought zero shot CoT self consistency tree thought ReAct chaining automatic meta, parameters
temperature top p top k presence frequency max tokens stop sequences system user assistant
context window overflow injection jailbreak PII bias toxicity hallucination grounding RAG vector DB
embedding Pinecone chunking reranking evaluation BLEU ROUGE BERTScore perplexity human
LLM as judge cost latency optimization compression caching fine tuning LoRA business application
customer support content code summarization classification extraction translation Q&A; safety RLHF
filtering ethical data leakage adversarial versioning logging monitoring observability cost rate latency
caching tool function calling agents multi agen...
Question 4: [D685OA-Q004] PROMPT COMPONENTS INSTRUCTION CONTEXT INPUT OUTPUT
INDICATOR EXAMPLES - Application: Student Amara WGU, AI Lab Houston, safety ethics.
Safety: PII redaction, bias check. Per practical prompt engineering what occurs?
A. Prompt components instruction context input output indicator examples - Correct: prompt components
instruction context input output indicator examples applied to safety ethics per D685
B. Distractor: Violates Prompt components instruction context input output indicator examples - misapplies
prompting or ignores safety for safety ethics
C. Distractor: Confuses Prompt components instruction context input output indicator examples with similar LLM
concept, misreads Safety: PII redaction, bias check
D. Distractor: Claims Prompt components instruction context input output indicator examples not needed for
prompt engineering safety ethics
EXAMINATION DOCUMENT 2026/2027
WGU D685 OA Final Exam
Practical Applications of Prompt Engineering
Objective Assessment Complete Questions & Detailed Verified Answers
Original Study Guide - Different Look & Different Concept Per Question - Other Format Reused
,Question 1: [D685OA-Q001] PROMPT ENGINEERING CASE: prompt design patterns at Prompt
Engineering Center - Dr Jonah AI notes: Prompt: Few-shot with chain-of-thought. Focus: Prompt
engineering definition objective subjective assessment WGU D685. Choose best D685 answer.
A. Distractor: Violates Prompt engineering definition objective subjective assessment WGU D685 - misapplies
prompting or ignores safety for prompt design patterns
B. Prompt engineering definition objective subjective assessment WGU D685 - Correct: prompt engineering
definition objective subjective assessment wgu d685 applied to prompt design patterns per D685
C. Distractor: Confuses Prompt engineering definition objective subjective assessment WGU D685 with similar
LLM concept, misreads Prompt: Few-shot with chain-of-thought
D. Distractor: Claims Prompt engineering definition objective subjective assessment WGU D685 not needed for
prompt engineering prompt design patterns
Correct answer: B
Detailed Verified Rationale (Prompt engineering definition objective subjective assessment WGU
D685): Prompt case prompt design patterns at Prompt Engineering Center, Dr Jonah AI documents
Prompt: Few-shot with chain-of-thought. Prompt engineering definition objective subjective
assessment WGU D685 per WGU D685 OA Final Exam 2026/2027 Practical Applications of Prompt
Engineering Objective Assessment covers prompt engineering definition LLM fundamentals
transformer tokenization, prompt components instruction context, zero one few shot role chain of
thought zero shot CoT self consistency tree thought ReAct chaining automatic meta, parameters
temperature top p top k presence frequency max tokens stop sequences system user assistant
context window overflow injection jailbreak PII bias toxicity hallucination grounding RAG vector DB
embedding Pinecone chunking reranking evaluation BLEU ROUGE BERTScore perplexity human
LLM as judge cost latency optimization compression caching fine tuning LoRA business application
customer support content code summarization classification extraction translation Q&A; safety RLHF
filtering ethical data leakage adversarial versioning logging monitoring observ...
Question 2: [D685OA-Q002] Select BEST answer: Large language model LLM fundamentals
transformer architecture - AI Solutions Austin - LLM evaluation. Dr Priya LLM - Model: GPT-4
temperature 0.7 top-p 0.9.
A. Best: Large language model LLM fundamentals transformer architecture - Correct complete verified answer
per WGU D685 OA Final
B. Distractor: Violates Large language model LLM fundamentals transformer architecture - misapplies prompting
or ignores safety for LLM evaluation
C. Distractor: Confuses Large language model LLM fundamentals transformer architecture with similar LLM
concept, misreads Model: GPT-4 temperature 0.7 top-p 0.9
D. Distractor: Claims Large language model LLM fundamentals transformer architecture not needed for prompt
engineering LLM evaluation
Correct answer: A
, Detailed Verified Rationale (Large language model LLM fundamentals transformer architecture):
Prompt case LLM evaluation at AI Solutions Austin, Dr Priya LLM documents Model: GPT-4
temperature 0.7 top-p 0.9. Large language model LLM fundamentals transformer architecture per
WGU D685 OA Final Exam 2026/2027 Practical Applications of Prompt Engineering Objective
Assessment covers prompt engineering definition LLM fundamentals transformer tokenization, prompt
components instruction context, zero one few shot role chain of thought zero shot CoT self
consistency tree thought ReAct chaining automatic meta, parameters temperature top p top k
presence frequency max tokens stop sequences system user assistant context window overflow
injection jailbreak PII bias toxicity hallucination grounding RAG vector DB embedding Pinecone
chunking reranking evaluation BLEU ROUGE BERTScore perplexity human LLM as judge cost
latency optimization compression caching fine tuning LoRA business application customer support
content code summarization classification extraction translation Q&A; safety RLHF filtering ethical data
leakage adversarial versioning logging monitoring observability cost rate latency cac...
Question 3: [D685OA-Q003] Which is TRUE about Tokenization tokens token count cost
optimization? Prompt case at LLM Research Dallas, RAG, Dr Carlos data Evaluation: BLEU 0.78
ROUGE-L 0.82. WGU D685 OA Final.
A. Distractor: Violates Tokenization tokens token count cost optimization - misapplies prompting or ignores
safety for RAG
B. Distractor: Confuses Tokenization tokens token count cost optimization with similar LLM concept, misreads
Evaluation: BLEU 0.78 ROUGE-L 0.82
C. Correct for Tokenization tokens token count cost optimization: detailed verified per WGU D685 OA
2026/2027 practical prompt engineering
D. Distractor: Claims Tokenization tokens token count cost optimization not needed for prompt engineering RAG
Correct answer: C
Detailed Verified Rationale (Tokenization tokens token count cost optimization): Prompt case RAG at
LLM Research Dallas, Dr Carlos documents Evaluation: BLEU 0.78 ROUGE-L 0.82. Tokenization
tokens token count cost optimization per WGU D685 OA Final Exam 2026/2027 Practical Applications
of Prompt Engineering Objective Assessment covers prompt engineering definition LLM fundamentals
transformer tokenization, prompt components instruction context, zero one few shot role chain of
thought zero shot CoT self consistency tree thought ReAct chaining automatic meta, parameters
temperature top p top k presence frequency max tokens stop sequences system user assistant
context window overflow injection jailbreak PII bias toxicity hallucination grounding RAG vector DB
embedding Pinecone chunking reranking evaluation BLEU ROUGE BERTScore perplexity human
LLM as judge cost latency optimization compression caching fine tuning LoRA business application
customer support content code summarization classification extraction translation Q&A; safety RLHF
filtering ethical data leakage adversarial versioning logging monitoring observability cost rate latency
caching tool function calling agents multi agen...
Question 4: [D685OA-Q004] PROMPT COMPONENTS INSTRUCTION CONTEXT INPUT OUTPUT
INDICATOR EXAMPLES - Application: Student Amara WGU, AI Lab Houston, safety ethics.
Safety: PII redaction, bias check. Per practical prompt engineering what occurs?
A. Prompt components instruction context input output indicator examples - Correct: prompt components
instruction context input output indicator examples applied to safety ethics per D685
B. Distractor: Violates Prompt components instruction context input output indicator examples - misapplies
prompting or ignores safety for safety ethics
C. Distractor: Confuses Prompt components instruction context input output indicator examples with similar LLM
concept, misreads Safety: PII redaction, bias check
D. Distractor: Claims Prompt components instruction context input output indicator examples not needed for
prompt engineering safety ethics
