Design an enterprise AI assistant like glean
PremiumLLM systemMedium
- Permission aware retrieval
- RAG
- ACL
Quick answer
This LLM system interview question asks you to design an enterprise AI assistant like glean. Aim for a design that uses language models to serve users well, then explain retrieval, prompting, serving, evaluation and trade-offs. Interviewers want to see how you control quality, cost and latency, and how you evaluate the results.
The problem
Design an enterprise AI assistant like Glean. Employees ask questions and get cited answers drawn from company documents, chat messages and tickets across many apps. The hard parts are permission-aware retrieval, keeping 50 million items fresh, and never showing an employee content they cannot open in the source app.
The assistant is sold to many companies. Each company is a separate tenant; design for the largest one.
What must an enterprise AI assistant like glean do?
- Permission-aware results: Use only items the asking employee can open in the source app; results, answers and citations all follow this rule.
- Cited answers: Answer in plain text with citations that link to the source items.
How well must it do it?
- Access control: An item the employee cannot open must never appear; permission removals take effect within 5 minutes.
- Latency: Search results within 500 ms at p95; first answer token within 2.5 s at p95.
What scale should you design for?
Use these numbers in your estimates.
The scale numbers unlock with this questionUnlock
Key concepts
- Permission aware retrieval
- RAG
- ACL
- Hybrid search
- Incremental sync
- Citations
How you’ll practise
- Problem framing
- Data and retrieval
- Prompting and orchestration
- Evaluation and safety
- Serving, cost and latency
- Wrap-up
Lala, personalized to you, gives a hint when you ask and reviews your design when you are ready.
What do interviewers look for?
In any LLM system design round.
- Defines the user task and what a good answer is
- Explains how relevant context reaches the model
- Plans evaluation before and after launch
- Controls cost and latency with caching and routing
- Covers safety, privacy and abuse
- Handles failures and fallbacks gracefully
Questions people ask about this interview question
- Which design type is this question?
- It is an LLM system question. You design an enterprise AI assistant like glean and explain your choices the way you would in an interview.
- Is this question asked at entry level?
- It is marked Medium. Entry-level engineers can practise it, but it expects you to reason about requirements and trade-offs, not only list components.
- What should I cover first?
- Start with the user task, what a good answer looks like and how it will be measured. Agree on scope with the interviewer before you design anything.
- Can I practise it for free?
- It is a premium question. Sign up and unlock it once for 10 credits; it then stays unlocked. When you finish, Lala, your personalized AI mentor, can review your design for 30 credits.
Related questions
Updated · Written by the SevyDevy team