Design a RAG Chatbot (GenAI Roles)
An Advanced Applications Engineering Analyst candidate got that prompt in June 2026, in a 15 to 20 minute campus interview that "revolved around the basics of GenAI" and the projects on their resume. Your client may sell something else. The shape is what repeats. An AIML Engineer in Bengaluru (August 2026) was asked, face to face at an Accenture center, about "developing an architecture for a GenAI project," including how to evaluate RAG with a graph database. An AI LLM Engineer (April 2026) described a technical round on "LLM architecture, RAG pipelines, and system design." So this lesson builds the pharmacy bot end to end, because a concrete client makes every choice easier to defend. The same design answers the other two prompts. RAG stands for retrieval-augmented generation: before the model writes an answer, the system finds the passages in the client's own documents that answer the question and puts them in the prompt. The model then writes from those passages instead of from memory.
AccenturePremium lesson · System design round
The Accenture design, worked end to end.
- Requirements and scopeSystem design · Frequently asked
- The API and data model
- The design, drawn out1 diagram
- 6 deep divesWith the tradeoffs
$49 a year · all 107 companies · instant access
20,000+ candidate reports
Accenture interview guideKnow what Accenture asks before you walk in.
23 coding · 5 system design · 34 behavioral questions
Get their questions first.
InterviewLegend is independent and not affiliated with or endorsed by any company it covers.