SaveMyToken logoSaveMyToken.Coding Plan
← Learning library

FREE AI INTERVIEW QUESTIONS & ANSWERS

AI Interview Questions and Answers: Free Topic Practice

Prepare for AI engineer and Forward Deployed Engineer (FDE) interviews with free topic questions, matched source answers, foundation practice, and customer scenarios. All questions and answers are in English, with original source numbers preserved.

Every answer is free No sign-up No API key
604 free questions10 technical topics24 FDE scenarios

AI fundamentals: questions and answers

Explain the concept aloud, reveal the reference answer, and use the linked lab to try it yourself.

10 questions · Showing 1–10

Beginner · Vector search02

When do dot product and cosine give the same ordering?

Reveal reference answer

For vectors normalized to unit length, their dot products equal their cosine similarities. Otherwise vector magnitude can change dot-product ranking.

Watch for: Do not normalize blindly if the embedding model expects a different scoring method.

Try the lab: For Nonzero Vectors, Cosine Similarity Depends on Direction ↗
Beginner · Vector search03

Is a similarity score a probability of correctness?

Reveal reference answer

No. It is a score under a particular representation and metric. Calibrate retrieval decisions with labeled tasks and verify answer claims separately.

Watch for: A high semantic match can retrieve an outdated or factually wrong passage.

Try the lab: For Nonzero Vectors, Cosine Similarity Depends on Direction ↗
Beginner · Vector search05

How should you choose an embedding model?

Reveal reference answer

Evaluate representative language, document types, queries, and relevance labels. Compare retrieval quality with latency, storage, and total processing cost.

Watch for: A larger dimension alone does not prove better results for your workload.

Try the lab: For Nonzero Vectors, Cosine Similarity Depends on Direction ↗
Intermediate · Vector search07

Why can filtering after Top K miss valid evidence?

Reveal reference answer

The global shortlist may contain only ineligible records. Filtering that truncated list cannot recover eligible records excluded before filtering.

Watch for: An empty shortlist does not prove the eligible corpus contains no answer.

Try the lab: Filtering After Top K Can Hide Eligible Vector Search Results ↗
Intermediate · Vector search08

How does exact search differ from approximate search?

Reveal reference answer

Exact search considers the full eligible set under the defined metric. Approximate search trades some retrieval accuracy for efficiency through an index or candidate-selection method.

Watch for: This three-record sorting example is not evidence of approximate-index performance.

Try the lab: Filtering After Top K Can Hide Eligible Vector Search Results ↗
Intermediate · Vector search09

How should updated documents be handled?

Reveal reference answer

Associate chunks with a document version, publish a consistent updated set, and remove or exclude superseded chunks. Test retrieval immediately after the change.

Watch for: Adding new vectors while keeping old conflicting passages active can mix policy versions.

Try the lab: Filtering After Top K Can Hide Eligible Vector Search Results ↗
Intermediate · Vector search10

What should you measure when selecting an index?

Reveal reference answer

Measure recall against an exact baseline together with latency, memory, build time, updates, and filtering behavior on representative data.

Watch for: A latency number without corpus size and recall conditions is hard to interpret.

Try the lab: Filtering After Top K Can Hide Eligible Vector Search Results ↗