LunaPromptsThe candidate opens the call. The interviewer opens a blank document. Within five minutes, they are asking about something the assessment already established. Both people lose a chance to explore the question that actually matters.
An interviewer copilot is most useful when it helps prepare that missing question. The surrounding process should preserve the answer so the following round can move forward.
LunaPrompts provides a dedicated interview environment with integrity signals and an AI copilot for human interviewers. It helps the team prepare from existing assessment evidence, investigate unanswered questions, and bring findings from multiple rounds into a combined candidate report.
Carry one question across the handoff
Human interviewerIllustrative scenario01 / 03Read the assessment gap
The API works, but its behavior after a lost response is unclear.
A candidate’s API assessment demonstrates correct validation and data handling. Their explanation of retries is brief. Before the interview, write a focused objective: understand whether the candidate can reason about duplicate operations after an ambiguous failure.
Open with the submitted code and a concrete change: the server completes the operation, but the response never reaches the caller. Ask what the caller should do and how the service would recognize a duplicate request. Discuss the tradeoff of the proposed approach at the scope expected for the role.
Leave room for the candidate to ask questions about the team. A tightly prepared technical conversation should create that room, rather than fill the call with more generated questions. The interviewer can follow a useful explanation without treating the copilot’s outline as a script.
Afterward, record which part of the uncertainty was resolved and which part remains open. Attach the conclusion to the response or work that supports it. In a later round, the next interviewer can explore a separate architecture or collaboration question instead of asking about retries again.
Before the interview: choose the questions that matter
The interviewer copilot helps prepare the conversation around the role and the candidate's earlier work. It can suggest topics, identify areas that were not adequately assessed, and help the interviewer decide where to spend their limited time.
A coding assessment might establish implementation ability without showing how a candidate handles unclear requirements. A design task might demonstrate architecture knowledge but leave a reliability tradeoff unexplored. Those gaps are useful starting points for the interview.
The interviewer should review the suggestions and select questions appropriate to the role. Preparation works best when it produces a small number of purposeful topics rather than a long script.
During the interview: support the conversation
The interview environment combines the human conversation with AI Judge assistance and the integrity capabilities enabled for that session. The interviewer can use the available context to explore a candidate's answer more precisely.
Integrity indicators should be reviewed with the interview's tool policy and supporting evidence. A secured environment adds controls and visibility; it cannot guarantee that every answer is independent or correct.
The human interviewer remains responsible for the discussion, for checking important claims, and for giving the candidate a fair chance to explain their thinking.
After the interview: connect the rounds
The combined report is intended to bring together the evaluation stages used in your hiring workflow. These can include an AI screening conversation, an L0 phone interview, a practical assessment, and human L1, L2, or L3 rounds.
The names of those rounds vary by organization. What matters is preserving the evidence each round contributes. A final review should make it easier to see which requirements have been demonstrated, where reviewers agree, and what remains unresolved before an offer decision.
Avoid reducing a nuanced candidate to one unsupported recommendation. A useful combined report helps the hiring manager inspect the reasons behind the recommendation.
A review framework your team can use
The following is an editorial checklist for this workflow. Adapt it to the role, the candidate journey, and your configured environment.
| Review area | What to establish |
|---|---|
| Before | One objective drawn from existing evidence |
| During | A changed condition and a specific discussion |
| After | An evidence-linked note |
| Next round | A new question, not the same interview again |
A combined report should not flatten disagreement into a confident summary. Preserve material differences between reviewers and the work each person relied on.
Put the guide to work
Take a recent interview plan. Mark each question as already assessed, unresolved, or new. Replace one repeated question with a focused follow-up.
Download the editable worksheet to record the decisions and open questions with your team. The examples in this guide are illustrative, not customer results or validated selection rules. Product availability and setup depend on the workflow agreed for your account.
A focused checklist, a worked scenario, and questions to resolve together.
Frequently asked questions
What does the interviewer copilot do before a call?
It helps prepare role-relevant questions using available assessment and interview context, including areas the earlier evaluation did not fully cover.
Can a secure interview guarantee that the candidate is telling the truth?
No. Environment controls and integrity signals support review. The team still needs to examine the evidence and assess the candidate's explanations.
Does every company need L0 through L3 rounds?
No. Those are examples of stage names. Use only the rounds that answer a necessary hiring question, and make the process clear to candidates.
Keep building your hiring workflow
- Candidate analytics: The scores match. The follow-ups should not.
- AI interviews: More first conversations. A clearer handoff to people.
- Recruiting copilots: Give the agent a task. Give the exception a person.
Explore interview copilot with LunaPrompts or browse all 19 hiring field guides.
