Design a RAG-based assistant service

Q: Design a RAG-based assistant service

This question evaluates system-design and machine-learning engineering competencies related to Retrieval-Augmented Generation, including architecture for retrieval and indexing, access control and tenant isolation, freshness and observability, citation and hallucination mitigation, and safety/PII handling.

Q: How do I approach ML System Design interview questions?

ML System Design questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master ml system design interviews.

Question

Loading...

Scenario

You need to build a Retrieval-Augmented Generation (RAG) assistant for an enterprise product. It should answer questions using internal documents and return grounded answers with citations.

Task

Design the end-to-end RAG system.

Requirements

Multi-tenant (each enterprise/customer isolated).
Access control enforcement (document-level and snippet-level).
Freshness: new/updated docs searchable quickly.
Citations and low hallucination rate.
Observability and evaluation.

What to cover

Ingestion and indexing pipeline (chunking, embeddings).
Retrieval, reranking, and context assembly.
Generation strategy and citation mechanism.
Safety/PII handling.
Metrics and testing strategy.

Design a RAG-based assistant service

Overview

Scenario

Task

Requirements

What to cover

Comments (0)