1 min lesson
Self-scoring against the ADM bar
Connect "Self-scoring only works if you're honest about it" to the decision you would make.
Step 1 of 4
Four deliverables done. Now grade them the way the panel would - across five dimensions, with a hard look at anything you overclaimed.
Self-scoring only works if you're honest about it. The whole point of this capstone is to find your weak dimension while it's free to fix. Score each one 1–5, then read the consistency check, because a loop that contradicts itself across deliverables fails even if each piece looks strong alone.
Learn more
Advanced table
Truth-seeking is weighted heavily here
- Dimension
- Product depth
- A 5 looks like
- You speak Tab, Agent, context,
.cursor/rules/, MCPModel Context Protocol. A standard that lets an AI agent pull in context from outside the repo, like Jira tickets or internal docs. Press Enter for the full definition. fluently and tie each to a workflow outcome - A 2 looks like
- Generic “the AI helps developers” with no specific Cursor capability
- Dimension
- Rollout / operational rigor
- A 5 looks like
- Phased plan with metrics and leading indicators per phase, risks named with mitigations
- A 2 looks like
- A list of activities with no numbers and no sequencing logic
- Dimension
- Data reasoning
- A 5 looks like
- You segmented the stall and traced each action to a specific signal
- A 2 looks like
- You reached for “more training” against an aggregate number
- Dimension
- Executive communication
- A 5 looks like
- EBR opens in the sponsor's goals, pairs data with stories, surfaces renewal early
- A 2 looks like
- A dashboard read-out with no narrative the CTO would repeat
- Dimension
- Truth-seeking honesty
- A 5 looks like
- You refused to inflate ROIReturn on Investment. The value gained versus what it cost, the language an economic buyer funds deals in. Press Enter for the full definition. and flagged your own uncertainty
- A 2 looks like
- You promised a productivity multiplier you can't defend
| Dimension | A 5 looks like | A 2 looks like |
|---|---|---|
| Product depth | You speak Tab, Agent, context, .cursor/rules/, MCPModel Context Protocol. A standard that lets an AI agent pull in context from outside the repo, like Jira tickets or internal docs. Press Enter for the full definition. fluently and tie each to a workflow outcome | Generic “the AI helps developers” with no specific Cursor capability |
| Rollout / operational rigor | Phased plan with metrics and leading indicators per phase, risks named with mitigations | A list of activities with no numbers and no sequencing logic |
| Data reasoning | You segmented the stall and traced each action to a specific signal | You reached for “more training” against an aggregate number |
| Executive communication | EBR opens in the sponsor's goals, pairs data with stories, surfaces renewal early | A dashboard read-out with no narrative the CTO would repeat |
| Truth-seeking honesty | You refused to inflate ROIReturn on Investment. The value gained versus what it cost, the language an economic buyer funds deals in. Press Enter for the full definition. and flagged your own uncertainty | You promised a productivity multiplier you can't defend |
Truth-seeking is weighted heavily here. Cursor's hiring culture screens for it on purpose.
Interactive diagram. Tab through its regions; each focused region shows its detail in the panel below.
Score each 1–5 - but know which ones a Cursor loop weights hardest before you self-grade.