Security Engineering
AI/ML Security
7 practice questions. Free questions open a full answer guide; the rest unlock with Pro.
- Your team wants to send user data to a third-party LLM API. What security and privacy concerns would you raise before doing that?
- Your ML teams pull pretrained models and datasets from public hubs and fine-tune on them. Walk me through the supply-chain risks in that pipeline and how you'd contain them.Go Pro
- An autonomous AI agent in your product can call internal APIs and act on its own. What's the security risk you watch for most, and how do you bound what it can actually do?Go Pro
- In a Retrieval-Augmented Generation system, why is the retrieval corpus a security boundary, and how do you defend against an attacker poisoning the documents the model retrieves?Go Pro
- When an LLM agent can call tools or take actions, how do you apply least privilege to limit the damage from a model that's been manipulated or just behaves unexpectedly?Go Pro
- You want to use a pre-trained model or a public dataset you downloaded from the internet. What are the security risks, and how would you reduce them?Go Pro
- How would you defend an application that puts user input into an LLM prompt against prompt injection, and why isn't a 'don't follow malicious instructions' system prompt enough?Go Pro
Want questions matched to your role? Paste a job title, job description, or CV for a personalized set, or go Pro to unlock the full bank.