NLP Interview Questions

Foundational model question on BPE/tokenizers — why token counts differ from words, and production impact on cost, limits, and RAG chunking.

Vector databases power every RAG system, yet most candidates can't explain ANN algorithms or hybrid search. This fundamental question appears in 80% of AI engineering loops. Master dense vs sparse retrieval and when hybrid search wins.

500M vectors at sub-100ms p99 is a staff-level vector search design question from Uber, Airbnb, and large-scale ML platform teams. Learn sharding strategies, index tuning, and the operational trade-offs that separate senior from principal engineers.

Medium vector DB / embeddings trade-off question on dimensions vs cost/quality.

Medium RAG interview on hard negatives for embedding/reranker quality.

Hard RAG ops question on index versioning, dual-write, and blue-green cutovers.

**Hybrid** Traditional OCR for text layer + Gemini for semantic extraction.

**Model selection** Benchmark retrieval on your corpus — smaller models cheaper if recall OK.

**Fit** Nightly evals, backfill embeddings, document tagging — not interactive chat.

Latency, availability, freshness, recall, and error budget definitions with realistic dependencies.

Latency, recall proxies, ingest lag, index size, filter rates, error budgets, and RAG downstream signals.

Encryption at rest/transit, tenant isolation, ACL on metadata filters, PII in embeddings, and audit logging.

Learned sparse representations, inverted index integration, lexical expansion, and fusion with dense ANN.

efSearch, nprobe, over-fetch for filters, offline benchmarks, and continuous monitoring of retrieval quality.

Shared embedding spaces, separate indexes, CLIP-style models, metadata routing, and fusion strategies.

Dual-write, shadow traffic, embedding compatibility, cutover rollback, and validation gates.

Snapshot strategies, embedding model lineage, cross-region restore, and rebuild-from-source playbooks.

PCA, Matryoshka embeddings, learned compression, and recall impact when shrinking vector size.

Horizontal scaling patterns: hash sharding, semantic partitions, routing embeddings, and merge/rerank at query time.

Embedding geometry and index metric choice — normalization, MIPS, and provider defaults for text retrieval.

Embedding similarity, k-NN example banks, MMR diversity, and eval-driven example curation for few-shot prompts.

Medium RAG interview question on cost of embeddings at 100m documents — architecture, trade-offs, eval, and production patterns.

Hard RAG interview question on databricks interview: lakehouse rag — architecture, trade-offs, eval, and production patterns.

Hard RAG interview question on rag for structured + unstructured data — architecture, trade-offs, eval, and production patterns.