ather-techie/rag-interview-questions

This repository is a study guide for people preparing for job interviews where they might be asked about RAG, which stands for Retrieval-Augmented Generation. RAG is an approach to building AI systems where the AI looks up relevant information from a database before generating its answer, rather than relying solely on what it learned during training. It is a widely discussed topic in AI engineering roles. The guide covers 10 different varieties of RAG, each in its own section file. These range from the simplest form (Naive RAG, which just retrieves chunks of text and passes them to the AI) to more involved approaches like Agentic RAG (where the AI decides on its own when and how to fetch information), Graph RAG (which uses a knowledge graph to find related entities), Self-RAG (where the model is trained to evaluate and critique its own retrievals), and Multi-modal RAG (which can retrieve across text, images, tables, and audio, not just written content). Each section contains 5 interview questions with detailed answers, and each question is tagged with a difficulty level: Basic, Intermediate, or Advanced. The README lists 100 questions total across the 10 sections, though it describes each section as having 5 questions with answers, so some sections likely contain more material than the index suggests. The guide is aimed at two groups: people preparing to be interviewed for AI engineering or ML roles, and interviewers who want a structured set of questions to assess candidates. It also lists related tools and frameworks that come up alongside RAG in practice, such as vector databases like Pinecone and Chroma, embedding models, and evaluation frameworks like RAGAS. The actual question content lives in separate markdown files linked from the main README, so the README itself is primarily a navigation index.