Design a natural-language AEP Q&A assistant

Quick Overview

This interview question evaluates ML product requirements, data/labeling, modeling, serving architecture, evaluation, monitoring, and trade-offs in a realistic interview setting. A strong answer for Design a natural-language AEP Q&A assistant states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

Design a natural-language AEP Q&A assistant

Company: Adobe

Role: Machine Learning Engineer

Category: ML System Design

Difficulty: hard

Interview Round: Onsite

Design a natural-language assistant for Adobe Experience Platform (AEP) to help marketers: ( 1) answer questions about any product functionality strictly using public-facing AEP documentation; ( 2) answer questions about the customer’s own data stored in AEP (e.g., “How many datasets do I have?”, “What are the top 5 biggest audiences by population?”) via secure summarization. Specify the end-to-end architecture (document ingestion, indexing, retrieval, grounding/orchestration, LLM selection, prompt design), authorization and data-access controls for tenant-scoped data, guardrails for scope-limited Q&A with a default refusal message for out-of-domain queries, latency/SLA targets, privacy/PII handling and auditability, evaluation/monitoring strategy, and how the system handles example questions like authentication steps and creating rule-based audiences. Provide proposed APIs, schemas, and a deployment plan.

Quick Answer: This interview question evaluates ML product requirements, data/labeling, modeling, serving architecture, evaluation, monitoring, and trade-offs in a realistic interview setting. A strong answer for Design a natural-language AEP Q&A assistant states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

|Home/ML System Design/Adobe
Adobe logo
Adobe
Aug 9, 2025, 12:00 AM
hardMachine Learning EngineerOnsiteML System Design
5
0

Design a natural-language AEP Q&A assistant

System Design: Natural-Language Assistant for Adobe Experience Platform (AEP)

Goal

Design a natural-language assistant for AEP that helps marketers:

  1. Answer questions about AEP product functionality strictly using public-facing AEP documentation.
  2. Answer questions about the customer’s own data stored in AEP (e.g., "How many datasets do I have?", "What are the top 5 biggest audiences by population?") via secure summarization.

Requirements

Specify:

  • End-to-end architecture: document ingestion, indexing, retrieval, grounding/orchestration, LLM selection, prompt design.
  • Authorization and data-access controls for tenant-scoped data.
  • Guardrails for scope-limited Q&A and a default refusal message for out-of-domain queries.
  • Latency/SLA targets.
  • Privacy/PII handling and auditability.
  • Evaluation/monitoring strategy.
  • How the system handles example questions like authentication steps and creating rule-based audiences.
  • Proposed APIs, schemas, and a deployment plan.

Clarifying Questions to Ask Guidance

  • Clarify users, core use cases, read/write patterns, scale, latency, availability, and data retention.
  • State explicit assumptions before making sizing or architecture decisions.
  • Prioritize the functional path first, then address reliability, security, observability, and rollout.

What a Strong Answer Covers Guidance

  • A scoped requirements summary with concrete non-goals and success metrics.
  • ML-specific data, model, evaluation, serving, and monitoring choices.
  • Reasoned trade-offs among simple and scalable designs, including bottlenecks and failure modes.
  • A validation, monitoring, migration, and launch plan appropriate for the risk level.

Follow-up Questions Guidance

  • What breaks first at 10x traffic or data volume?
  • How would you degrade gracefully during dependency failures?
  • What metrics and alerts would prove the design is healthy after launch?

Submit Your Answer to Earn 20XP

Sign in to leave a comment

Loading comments...