Evaluation interview question · Most asked
What guardrails would you put around an LLM feature?
Summary
Input: keep secrets out of prompts, screen for injection and abuse, treat retrieved and tool content as untrusted. Capabilities: least-privilege tools with authorisation in code and approvals for consequential actions. Output: schema and content validation, policy checks, personal-data checks, escaping for the destination. Plus monitoring and rate limits.
Longer Answer
Each layer can fail, so they are combined. Measure guardrail errors in both directions: harmful content allowed and legitimate requests refused.
Key Points
- Input checks.
- Least-privilege capabilities.
- Output validation.
- Measure both error types.
Deep dive: LLM Evals and Guardrails
More most-asked Evaluation questions
- Evaluation leaf: What is a golden dataset?
- When do you use code graders, model graders or human graders?
- How do you decide whether a prompt change is an improvement?
- How do you build an evaluation set for an LLM feature?
- Evaluation leaf: What is faithfulness/groundedness?
- Evaluation leaf: Pros/cons of LLM-as-judge?
- Evaluation leaf: Offline vs online evaluation tradeoffs?