Available question text
Design a retrieval-augmented generation (RAG) system that provides real-time insights on purchase orders for 1 million users with 100 million purchase records, where 10% of customers generate 90% of orders. Discuss the end-to-end architecture including retrieval, generation, and time-based caching strategies with various caching algorithms for optimal performance under skewed load.
Related questions
Continue with questions that share topics or the same question type.
- Build and Evaluate a Minimal Embedding-Based RAG Pipeline
ML system design · ML System Design
- Add an LLM-backed recommendation API to an existing service with RAG and prompting
ML system design · ML System Design
- Build a Small Agent or RAG Tool with the Mistral API
ML system design · ML System Design
- Design a Jira+Confluence RAG assistant
ML system design · ML System Design
Sources · 1 reference
These records explain where the question appears. A company tag or candidate report does not guarantee a future interview question.
Hello Interview · 1 record
SAP
Design a RAG-based AI insights system for purchase orders
Reported Mar 9, 2026 · 1 source report
SENIOR
Source snapshot: Nov 12, 2026
View source