Curated · Mid-Level
Top 100 Mid-Level Interview Questions
Here are the top 26 AI interview questions for Mid-Level candidates, sorted by community quality. Each entry links to a detailed answer with explanations, hints, and source-grade follow-ups. Browse all Mid-Level questions →
Sorted by community quality · refreshed as new questions are published.
- 01Explain scaled dot-product attention.Short AnswerMedium·Qual 4.8
- 02What is the KV cache in transformer inference?FlashcardEasy·Qual 4.7
- 03What is RLHF, and why is it used after pretraining?Multiple ChoiceMedium·Qual 4.7
- 04What does LoRA do, and why is it popular for fine-tuning?Multiple ChoiceMedium·Qual 4.7
- 05Which metric best measures whether a RAG answer is grounded in the retrieved context?Multiple ChoiceMedium·Qual 4.7
- 06When should you fine-tune instead of using RAG?Short AnswerMedium·Qual 4.6
- 07Explain why transformers replaced RNNs for language modeling.Short AnswerMedium·Qual 4.6
- 08What does RAG primarily help with in LLM-based applications?Multiple ChoiceEasy·Qual 4.6OpenAI
- 09Why combine BM25 with dense embeddings for retrieval?Multiple ChoiceEasy·Qual 4.6
- 10PPO vs DPO, what's the practical difference?FlashcardMedium·Qual 4.6
- 11What is the Model Context Protocol (MCP) and what problem does it solve?Multiple ChoiceEasy·Qual 4.6
- 12What distinguishes an AI agent from a plain LLM call?Multiple ChoiceMedium·Qual 4.5
- 13Which components make up a standard transformer block?Multiple ChoiceEasy·Qual 4.5
- 14HNSW vs IVF, when do you pick each for a production vector index?FlashcardMedium·Qual 4.5
- 15How does MCP differ from OpenAI-style function calling?FlashcardMedium·Qual 4.5
- 16Match each vector index type to its primary characteristic.Match PairsMedium·Qual 4.5
- 17Describe the typical agent execution loop.Short AnswerMedium·Qual 4.4
- 18Which of these are valid considerations when choosing a chunking strategy for RAG?Multi-selectMedium·Qual 4.4
- 19Estimate the per-call cost of a typical RAG chatbot using GPT-4o-mini.FlashcardMedium·Qual 4.4
- 20Why is cosine similarity preferred over Euclidean distance for text embeddings?FlashcardEasy·Qual 4.4
- 21Why does BPE tokenization use subwords instead of words or characters?FlashcardEasy·Qual 4.4
- 22Spot the error in this explanation of temperature.Spot the ErrorEasy·Qual 4.3
- 23Put the steps of a basic RAG query pipeline in order.Order StepsEasy·Qual 4.3
- 24What is semantic chunking and when does it beat fixed size chunking?FlashcardMedium·Qual 4.3
- 25Fill in the scaled dot-product attention formula.Fill in BlankEasy·Qual 4.2
- 26Predict the output: how many tokens does this string become with a typical BPE tokenizer?Predict OutputEasy·Qual 4.0
26 of 100 requested: only 26 published questions tagged for Mid-Level so far.
Related top picks
Change experience level
By topic
Top 100 Inference Optimization Interview Questions→Top 100 Fine-Tuning Interview Questions→Top 100 LLM Evaluation Interview Questions→
By role