Build the PRD for an Internal RAG Knowledge Assistant
Overview
What this challenge is about.
You receive: a description of the CS workflows (post-sale onboarding, escalation, renewal), an inventory of internal knowledge sources (Notion, Salesforce, Zendesk macros, 3 product wikis), and the previous failed attempt's post-mortem. Write a 6-page PRD covering: scope, user stories, ranked feature list, evaluation plan (offline + online), and explicit non-goals. Define 4 launch-quality metrics with target thresholds (e.g., faithfulness, response time, citation accuracy, deflection rate). Deliver a 1-page critical-path risk register.
The Brief
What you'll do, and what you'll demonstrate.
Write a defensible PRD for a RAG knowledge assistant that the CS org can hold engineering accountable to.
Earning criteria — what you'll demonstrate
- Scope a RAG product with explicit goals and non-goals
- Design evaluation that distinguishes offline (faithfulness) from online (deflection) signals
- Reason about the failure modes specific to RAG (hallucination, stale data, retrieval miss)
- Communicate AI-product trade-offs to a non-engineering VP
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
Careers
Roles this prepares you for.
Real titles. Real skill bridges. Pick the one closest to your trajectory.
AI Product Manager
Writing a defensible AI-product PRD with explicit goals, non-goals, and evaluation is the canonical AI PM portfolio piece.
This challenge sharpens
- product-management
- prd-writing
- evaluation-design
AI Product Designer
PRD-quality scoping with user-flow diagrams is the bridge into AI product designer roles at companies that pair PM + design tightly.
This challenge sharpens
- product-management
- stakeholder-analysis
- prd-writing