Skip to content
calwm

Red pen: Fundamentals of GenAI

Exam answers on Fundamentals of GenAI are often almost right. Each round hands you a short document with a few mistakes hidden in it. Mark the faulty lines, say why each one is wrong, and see what a good version would say.

Find the mistake before the auditor does.

Grab the red pen2 documents, about 4 min

What you will practise

You will practise Fundamentals of GenAI, with questions touching GenAI capabilities and GenAI concepts.

  • GenAI capabilities
  • GenAI concepts

It is part of the AWS Certified AI Practitioner track (AIF-C01), where Fundamentals of GenAI is about 24% of the exam.

A taste of what you will get

  • Leo needs replies in under 2 seconds that match Shelfline's returns policy, within a fixed budget.

    Sample document 1: Mei's launch checklist for the support assistant. Which lines are wrong, and why?

    • Purpose: answer customer questions about orders and returns
    • Model: the largest model available, because bigger is always better
    • Knowledge: answers come from what the model learned in training
    • Quality: Nora reviews a sample of answers every week
    • Escalation: customers can ask for a person at any time

    Answer: Line 2 (Model: the largest model available, because bigger is always better): A smaller model may meet the 2-second target for less. Better: Test smaller models first, and pick the smallest one that meets the speed, quality and budget limits. Line 3 (Knowledge: answers come from what the model learned in training): It can't know Shelfline's policy, so it may make answers up. Better: Retrieve the current policy documents for each question (RAG) so answers are grounded in them.

  • Sam must keep the assistant within budget at 40,000 chats a month.

    Sample document 2: Sam's cost plan for the support assistant. Which lines are wrong, and why?

    • Pricing: on-demand, so there's no monthly commitment
    • Only the model's replies are billed, so the long policy in every prompt is free
    • Index each 40-page policy as one single chunk in the knowledge base
    • Track cost per chat each week against the budget
    • Cap the reply length so answers stay short

    Answer: Line 2 (Only the model's replies are billed, so the long policy in every prompt is free): Input tokens are billed too, so it costs on every chat. Better: Count input and output tokens, and retrieve only the relevant policy passages. Line 3 (Index each 40-page policy as one single chunk in the knowledge base): Smaller chunks let retrieval find the right passage. Better: Split the policies into smaller chunks so retrieval returns only the relevant passage.

Advertise here

How scoring works

  • Every correct answer pays 10 XP for each difficulty level, so an easy item is worth 10, a medium one 20 and a hard one 30.
  • Answer with time to spare and you can earn up to 5 bonus XP on top.
  • A wrong answer pays nothing. It does not cost you anything either.
  • Flashcards pay 2 XP each time you grade yourself 3 or higher.
  • Using a hint costs 2 XP, so use it when it helps, not when it is comfortable.
  • Finish a session on consecutive days to build a streak. Your progress lives in your browser, and signing in syncs it.

Quick answers

How long does the red pen on Fundamentals of GenAI take?
About 4 minutes for 2 documents. Pause whenever you like.
Do I need an account to play the red pen?
No. Progress is saved in your browser, and signing in syncs it across devices.