What to collect
- A focused patch with a reproducible regression test
- Notes explaining a suggestion that required correction
- A short walkthrough of edge cases and remaining limitations
AI-ASSISTED DEVELOPMENT
Assess coding with the tools people actually use
Allow the tools the role needs. Evaluate the candidate’s review, debugging, and judgment alongside the code they deliver.
01 / THE TASK
A practical starting point.
Give the candidate a small repository with a pagination bug and an incomplete test suite. Ask them to fix the bug, add regression tests, and explain an AI-generated change they accepted or rejected. Agree the available editor and assistant before the session.
Adapt this example to the role, agree allowed tools, and give candidates the same instructions. It is a starting template, not a validated selection test.
02 / THE RUBRIC
Agree what good looks like.
Calibrate these suggested weights with your interviewers before the first candidate. Record the evidence behind each rating.
| Criterion | Weight | Stronger evidence | Needs investigation |
|---|---|---|---|
| Working change | 30% | Fixes the bug without breaking the existing behavior. | Handles the example but fails boundary cases. |
| Tests and verification | 30% | Demonstrates a failing test before the fix and meaningful regression coverage. | Accepts generated tests without checking what they assert. |
| Review of AI output | 25% | Spots an incorrect assumption and corrects it deliberately. | Cannot explain why the generated change is safe. |
| Communication | 15% | Explains the patch and unresolved tradeoffs in their own words. | Repeats the assistant’s explanation without connecting it to the code. |
03 / ILLUSTRATIVE REPORT
From observation to follow-up.
Example only · not a real candidate resultObserved
The sample patch passes its tests, but a generated test mocks away the boundary condition that caused the bug.
Reviewer’s next step
Ask the candidate to rewrite that test and explain the missing signal. Completion speed alone is insufficient evidence.
Ask about their own work.
- Which assistant suggestion did you reject, and why?
- Show the test that would fail if this bug returned.
- What happens when the input is empty or the last page is full?
04 / SEE THE PRODUCT
Bring the brief to your demo.
This LunaPrompts product walkthrough introduces the assessment workflow. Ask the team to demonstrate your exact task, chosen tools, and report requirements in the environment you plan to use.
The walkthrough shows the product. It is not a recording of the illustrative exercise above or an independent performance benchmark.
Read the detailed workflow guide ↗PUT IT TO WORK
Start with the role you’re hiring for.
Bring a job description, a few must-have skills, and your hiring timeline. We’ll help you shape the evaluation.
Talk to us ↗