AI 202 Assessment
Show that you can decide how an agent plans, reasons, evaluates and adapts
What is assessed
The assessment checks judgment, not vocabulary. It has five parts with the weights below.
| Part | Weight |
|---|---|
| Knowledge Check (12 questions on workflow vs. adaptive agent, goal, state and action, planning, ReAct, search and the MCTS concept, verification, reflection and re-planning) | 15% |
| Planning & Decision Scenarios (what is the goal, what state matters, what actions are possible, what evidence is needed, what next step to select) | 20% |
| ReAct / Search Exercises (trace Reason, Act, Observe, update the state, compare candidate actions, choose the next action) | 20% |
| Evaluation & Re-planning Exercises (Verify, Reflect, Continue, Re-plan, Escalate or Stop) | 15% |
| Final Application (Agent Planning & Reasoning Canvas) | 30% |
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, initial state and observations | 10% |
| Subgoals, available actions and tools | 15% |
| Planning strategy and decision policy | 20% |
| Evaluation and reflection | 15% |
| Re-planning triggers | 15% |
| Constraints, human escalation and stop conditions | 15% |
| Final output and reflection question | 10% |
- 1Excellent
The goal has success criteria and limits. Subgoals and actions are clear and irreversible or high-impact actions are marked. The strategy fits the task and each decision has options and criteria. Every result has an evaluation method, reflection has a limit, re-planning triggers are specific, and human escalation and stop conditions are placed where the risk is. - 2Competent
The main plan and actions are designed, some decisions lack criteria or some triggers are vague, and the human controls are broadly sensible. - 3Needs improvement
The goal or actions are vague, the agent is left to decide everything, results are never evaluated, there are no re-planning triggers or stop conditions, or high-impact actions have no human approval.
Reflection
Write 100 to 200 words in response to this question:
Info
Why should this agent choose this next action, and what evidence would cause it to change its plan?
Start
- Start the AI 202 assessment (post-test)
- My Learning: Progress and results