Skip to content
calwm

Scenario: Fundamentals of GenAI

Real problems do not come with the facts lined up. You get a messy situation, a few decisions to make, and feedback on each one.

Make the call. Live with the outcome.

Start the scenario6 scenarios, about 12 min

What you will practise

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

  • AWS GenAI services
  • 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

  • Shelfline, a 300-person online homewares retailer, has launched its first AI support assistant. Sam, the finance analyst, flags that this month's AI bill has tripled, yet the number of customer chats is the same. Mei, the engineering lead, explains that every request now includes the full 40-page returns policy so the assistant always has it to hand.

    Same number of chats, a 40-page policy in every request, and triple the bill. Why?

    • Per-seat pricing: the policy counts as extra assistant users
    • Per-request pricing: the policy turns each chat into several requests
    • Per-hour pricing: longer requests keep the model running for longer
    • Per-token pricing: the policy adds thousands of tokens to each request

    Answer: Per-token pricing: the policy adds thousands of tokens to each request. Models are billed on tokens: the input they read and the output they write. Pasting 40 pages into every request multiplies input tokens even when the number of chats stays flat. Sending only the relevant part of the policy cuts the bill. Takeaway: you pay for every token in and out, so prompt length is a cost decision, not just a quality one.

  • Mei wants the assistant to do more than answer questions. It should look up a customer's order in the order system and log a ticket in the helpdesk tool. Rather than writing custom glue code for every tool, she wants one standard way to connect the assistant to these systems, which other tools could reuse later.

    One standard way to connect an AI assistant to outside tools and systems. Which fits?

    • Embeddings of the order and helpdesk data
    • Model Context Protocol (MCP)
    • Fine-tuning the model on order and ticket data
    • Retrieval Augmented Generation (RAG)

    Answer: Model Context Protocol (MCP). MCP is an open standard for connecting AI applications and agents to external tools and data sources. Each system is exposed once through MCP, and any compatible assistant can use it, with no custom glue per tool. Takeaway: MCP is how agents reach tools and data; it's the plumbing between the model and your systems.

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 scenario on Fundamentals of GenAI take?
About 12 minutes for 6 scenarios. Pause whenever you like.
Do I need an account to play the scenario?
No. Progress is saved in your browser, and signing in syncs it across devices.