← Anthropic · Claude Certified Architect – Foundations
Practice: CCAR-F
Question 1 of 8
Easy
When designing a prompt for a large language model, what is the primary benefit of providing few-shot examples?
Question 2 of 8
Hard
An application sends user documents to an LLM and must minimize the risk of prompt injection from document content. Which design is most effective?
Question 3 of 8
Easy
What does the 'temperature' parameter control in LLM text generation?
Question 4 of 8
Medium
A team wants an LLM application to answer questions using their private knowledge base without retraining the model. Which architecture fits best?
Question 5 of 8
Medium
Which practice most directly reduces hallucinations in an LLM application that answers factual questions?
Question 6 of 8
🔒 Premium
In a tool-use (function calling) architecture, what is the safest way …
Question 7 of 8
🔒 Premium
What is the purpose of a system prompt in a chat-based LLM application…
Question 8 of 8
🔒 Premium
A production LLM feature must keep per-request costs predictable. Whic…