AI 104 Assessment
Show that you can decide how an agent should execute a task step by step
What is assessed
The assessment checks judgment, not vocabulary. It has four parts with the weights below.
| Part | Weight |
|---|---|
| Knowledge Check (10 questions on prompt vs. workflow vs. agent, task decomposition, patterns, decision points, agent loop, verification and human-in-the-loop) | 20% |
| Workflow Pattern Decisions (scenarios: Sequential, Parallel + Combine, Routing, Loop) | 20% |
| Execution & HITL Decisions (scenarios: Fixed or Agentic? AI or Human? Continue or Escalate? Verify automatically or request approval?) | 20% |
| Final Application (Agentic AI Workflow Canvas) | 40% |
Scoring and passing
Quiz scoring happens on the server. Correct answers and explanations never appear on public pages or in downloadable files. The passing threshold is 80%.
Info
Questions appear once you start the assessment.
Canvas rubric
| Criterion | Weight |
|---|---|
| Goal, trigger and input clarity | 10% |
| Task decomposition (steps, inputs, outputs, dependencies, who executes) | 15% |
| Workflow pattern and execution order | 15% |
| Decision points (fixed, conditional, agentic, human) | 15% |
| Memory, knowledge, skills and tool steps | 15% |
| Verification and human-in-the-loop | 15% |
| Failure paths, fallback and final output | 15% |
- 1Excellent
Every step has an objective, input, output and executor. The pattern matches the dependencies, each decision point is classified with bounded options, tool steps are named, every result has a verification method, and human review, approval and fallback are placed where the risk is. - 2Competent
The main workflow is designed, some steps lack outputs or verification, and the human controls are broadly sensible. - 3Needs improvement
Steps are vague or everything is left to the agent, dependencies are missing, decisions are not classified, results are never checked, or high-impact actions have no human approval and no fallback.
Reflection
Write 100 to 200 words in response to this question:
Info
Which steps should be fixed, which should be agentic, and which should require human control?
Start
- Start the AI 104 assessment (post-test)
- My Learning: Progress and results