All 40 questions
What are embeddings, how do they work, and why are they used?
Asked in 9 of 191 interviews · last asked Oct 23 · Concept · Junior
Show answer
Embeddings are lists of numbers that represent text, images, or other data. A trained model creates them so that similar items tend to have similar vectors. They help systems find related items even when the words differ. In RAG, embeddings help retrieve relevant documents for the model to use.
- Training teaches the model which patterns and relationships to represent.
- Vector search compares embeddings using measures such as cosine similarity.
- Queries and documents need embeddings from the same model or compatible models.
- Embeddings support semantic search, recommendations, and clustering, but similarity does not guarantee relevance.
What the interviewer checks: The interviewer checks whether you understand how learned vectors capture similarity and support retrieval in RAG.
Other ways it was asked
- “What is your understanding of embeddings?”
- “What are embeddings, and when are they used?”
- “What are embeddings, and where are they used?”
How do you choose an embedding model for RAG and decide which documents to embed?
Asked in 4 of 191 interviews · last asked Nov 18 · Concept · Senior
Show answer
I choose an embedding model by testing retrieval on real questions with known relevant documents. I compare language and domain fit, cost, speed, and hosting needs. I embed clean, current, trusted content that users need, usually split into chunks, and remove duplicates and irrelevant material. I keep access permissions with each chunk and enforce them during retrieval.
- Compare models such as E5, BGE, and OpenAI embeddings on your own retrieval tests.
- Use compatible query and document encoders, following the model’s required input format.
- Choose chunk sizes that preserve meaning and fit the model’s input limit.
- Changing embedding models usually requires embedding the documents again.
What the interviewer checks: The interviewer checks whether you can test model choices and select useful, safe content for retrieval.
Other ways it was asked
- “Which embedding models do you know for RAG?”
- “How do you choose an embedding model for a RAG system, what documents should be embedded, and why?”
- “How would you choose an embedding model for a document retrieval system?”
What is cosine similarity, and how does it work on two vectors?
Asked in 4 of 191 interviews · last asked Oct 20 · Concept · Junior
Show answer
Cosine similarity measures how closely two nonzero vectors point in the same direction. It is the cosine of the angle between them: cos_sim(a, b) = (a · b) / (||a|| × ||b||). Compute their dot product, then divide by the product of their lengths.
- Scores range from -1 to 1: 1 means same direction, 0 means perpendicular, and -1 means opposite directions.
- It compares direction, not magnitude; multiplying a vector by a positive number does not change the score.
- It is undefined for zero vectors because the denominator is zero.
What the interviewer checks: The interviewer checks whether you understand the formula, its geometric meaning, and the zero-vector edge case.
Other ways it was asked
- “What is cosine similarity?”
- “What do you mean by cosine similarity?”
- “How does cosine similarity work on two numeric vectors?”
How would you create a vector and use it for semantic search?
Asked in 2 of 191 interviews · last asked Aug 19 · Write code · Mid-level
Show answer
I would use an embedding model to turn each document into a vector that represents its meaning. I would embed the search query with the same model, compare it with the document vectors, and return the closest matches. For large collections, I would use a vector index to make search faster.
- Use the same embedding model for documents and queries.
- Cosine similarity compares vector direction; dot product gives the same score when vectors have unit length.
- Vector indexes can speed up search, sometimes at the cost of missing relevant matches.
Python# Requires: pip install sentence-transformers from sentence_transformers import SentenceTransformer model = SentenceTransformer("all-MiniLM-L6-v2") documents = ["Dogs make good pets.", "Baking bread takes time.", "Cats are household companions."] vectors = model.encode(documents, normalize_embeddings=True) query_vector = model.encode("Animals to keep at home", normalize_embeddings=True) scores = vectors @ query_vector for i in scores.argsort()[::-1][:2]: print(documents[i], float(scores[i]))What the interviewer checks: The interviewer is checking whether you can connect embedding creation, similarity scoring, and retrieval.
Other ways it was asked
- “How would you write code to create a vector and perform semantic search on it?”
- “How can vector embeddings be used for semantic search?”
How does vector search find and score embeddings similar to a query?
Asked in 2 of 191 interviews · last asked Sep 9 · Concept · Mid-level
Show answer
Vector search embeds the query using the same embedding model as the stored vectors. It finds nearby vectors using an exact scan or an approximate nearest neighbor index. It ranks results using a measure such as cosine similarity, dot product, or Euclidean distance.
- Cosine similarity compares vector direction; dot product also depends on vector length.
- Higher similarity means closer matches; lower distance means closer matches.
- Approximate indexes speed up search but may miss some of the closest vectors.
- Returned scores may transform the raw similarity or distance; check the database’s definition.
What the interviewer checks: The interviewer checks whether you understand query embedding, nearest neighbor search, scoring, and the speed-versus-accuracy trade-off.
Other ways it was asked
- “How does a vector search system calculate similarity scores between a query and stored embeddings?”
- “How does vector search work?”
How would semantic search find results for a word such as “good” in vector space?
Asked once · last asked Aug 19 · Concept · Mid-level
Other ways it was asked
- “How does semantic search find results for a word such as “good” in vector space?”
Can semantic search be performed without embeddings?
Asked once · last asked Aug 19 · Concept · Mid-level
Other ways it was asked
- “What approach would you use to implement semantic search without embeddings?”
Which library would you use to calculate cosine similarity and match items?
Asked once · last asked Aug 19 · Tooling · Mid-level
How would you optimize embedding generation, and would batching improve throughput?
Asked once · last asked Jun 12 · Scenario · Senior
How can a multimodal RAG system represent text and images in a common format so both can be searched?
Asked once · last asked Jul 16 · Scenario · Senior
How can the simplifying assumptions of the vector-space model be addressed, particularly with the generalized vector-space model?
Asked once · last asked Aug 3 · Concept · Senior
Will an embedding model always produce the same embedding for the word “good”?
Asked once · last asked Aug 19 · Concept · Mid-level
How would you implement a cosine similarity function using NumPy?
Asked once · last asked Aug 19 · Write code · Mid-level
Which libraries would you use to generate and work with embeddings?
Asked once · last asked Aug 19 · Tooling · Mid-level
What tools or libraries can be used to compare face embeddings?
Asked once · last asked Sep 7 · Tooling · Mid-level
What determines the values in an embedding vector?
Asked once · last asked Sep 9 · Concept · Mid-level
What process converts text into a numerical vector for semantic search?
Asked once · last asked Sep 15 · Concept · Junior
How can cosine similarity be used to search for text such as “hello world” when it operates on numeric vectors?
Asked once · last asked Sep 15 · Concept · Mid-level
Other ways it was asked
- “How can cosine similarity be used to search for text such as “my name is Nikhil” when it operates on numeric vectors?”
How would you choose an embedding model and determine the dimensionality and other key parameters for a vector index?
Asked once · last asked Sep 23 · Scenario · Senior
Other ways it was asked
- “What embedding model did you choose, and how did you decide on the dimensionality or other key parameters for your vector index?”
How do embedding model size and parameter count affect embedding quality and computational cost?
Asked once · last asked Oct 20 · Concept · Mid-level
How is text converted into vectors for a RAG system, and how do the embedding model and vector storage system differ?
Asked once · last asked Oct 22 · Concept · Senior
Other ways it was asked
- “How is text converted into vectors for a RAG system, and when might you use different models for embedding and storage?”
What are vectors in RAG, and what happens under the hood when you upload a document to a RAG application?
Asked once · last asked Nov 4 · Concept · Mid-level
Other ways it was asked
- “After the documents are converted into vectors in a RAG application, what happens next?”
How would you evaluate the trade-offs between latency, memory usage, and retrieval quality when quantizing a search encoder?
Asked once · last asked Oct 28 · Scenario · Senior
Other ways it was asked
- “What criteria would you use to preserve retrieval quality when quantizing a search encoder?”
How do you prevent an AI-generated document from duplicating existing knowledge?
Asked once · last asked Nov 6 · Scenario · Senior
How do you prevent duplicate knowledge when the same idea is described differently or in different languages?
Asked once · last asked Nov 6 · Scenario · Senior
How does a natural-language query get converted into a vector search that retrieves relevant chunks from a vector database?
Asked once · last asked Nov 17 · Concept · Mid-level
What are embeddings, and what distinguishes dense embeddings?
Asked once · last asked Oct 28 · Concept · Mid-level
How are documents converted into embeddings in a RAG pipeline?
Asked once · last asked Nov 4 · Concept · Mid-level
What happens under the hood when a knowledge base checks for semantic duplicates?
Asked once · last asked Nov 6 · Concept · Senior
What is CLIP, and what makes this useful for image search?
Asked once · last asked Nov 12 · Concept · Mid-level
What is a vector in AI?
Asked once · last asked Aug 19 · Concept · Junior
Other ways it was asked
- “What are vectors used for in AI?”
Why do we vectorize text into numerical representations?
Asked once · last asked Feb 5 · Concept · Junior
Which sentence similarity models do you know?
Asked once · last asked Mar 14 · Concept · Mid-level
What are tokenization and embeddings?
Asked once · last asked Apr 28 · Concept · Mid-level
How do embeddings help represent information in numerical form?
Asked once · last asked Jun 10 · Concept · Junior
What properties of embeddings make them useful for representing semantic similarity?
Asked once · last asked Jun 12 · Concept · Mid-level
What is similarity search?
Asked once · last asked Jun 12 · Concept · Junior
What is an embedding?
Asked once · last asked Aug 7 · Concept · Junior
Other ways it was asked
- “What is embedding?”
What does the embedding model do?
Asked once · last asked Sep 15 · Concept · Junior
How do you check whether knowledge already exists in a database?
Asked once · last asked Nov 6 · Scenario · Mid-level
More RAG interview questions
- All RAG interview questions
- RAG Retrieval and Reranking Interview Questions
- RAG Evaluation and Debugging Interview Questions
- RAG System Design and Agent Interview Questions
How this list is made
Every question here was asked by a real interviewer in a real job interview that a candidate ran MeetAssist in. After the call, the interviewer's questions are pulled from the transcript. Practice sessions, tests, exams and anything that was not a job interview are dropped.
Questions are rewritten so they make sense on their own: names of people, employers and clients are removed, and speech-to-text mistakes are fixed. Small talk, logistics and questions about one candidate's own résumé are left out. Rephrasings of the same question are merged, and the count is the number of different interviews it came up in.
The interviews come from candidates who use MeetAssist, so they lean towards software, data and AI roles: the shares describe these interviews, not every RAG interview. Answers to the most asked questions are written and checked by an editor; a "common mistake" is shown only when reviews of real answers in these interviews show it, with the count.
191 interviews, 122 candidates, Nov 2025 to Nov 2026. Updated every month with new interviews.
Or add the Chrome extension