thinkhard.SI

Prompt repository Hiring & people

Decide which level to hire a candidate at from interview evidence

The panel wants to hire someone but disagrees on the level, and you need the evidence laid out against your level definitions before the hiring manager decides.

Registry no.
PPL-055
Collection
Hiring & people
Laid out for
Any chat model
Score
10 out of 10
Blanks to fill
8
Added

Hiring & peopleAny chat

The prompt

Score 10 of 10, Excellent

Use it when The panel wants to hire someone but disagrees on the level, and you need the evidence laid out against your level definitions before the hiring manager decides.

  • {{candidate_name}}
  • {{role}}
  • {{decision_maker}}
  • {{level_definitions}}
  • {{interview_evidence}}
  • {{levels_considered}}
  • {{pay_ranges}}
  • {{team_levels}}
Full prompt and why it works
Prompt
Lay out the interview evidence for {{candidate_name}}, a candidate for the {{role}} role, against our level definitions, so that {{decision_maker}} can decide which level to hire them at, and so that the decision, the offer and the candidate's future reviews rest on the same evidence. The panel has already agreed to hire; the open question is the level.

Our level definitions for this role:
"""
{{level_definitions}}
"""

Interview feedback and scorecards from the panel:
"""
{{interview_evidence}}
"""

Levels being considered: {{levels_considered}}
Pay range for each of those levels: {{pay_ranges}}
Where people already in the role sit, by level only: {{team_levels}}

Produce, with a heading for each part:
1. Evidence by level: a table with one row per criterion in the level definitions and the columns Criterion, Evidence for the lower level, Evidence for the higher level, Not tested. Quote or closely paraphrase the panel, with each interviewer's role.
2. The level the evidence best supports, with your confidence (strong, moderate or weak) and the pieces of evidence that decide it. This is a recommendation for {{decision_maker}}, not the decision.
3. Pay implications: where an offer would sit in each level's range, and how that compares with the levels of people already in the role.
4. Risks of each choice. For example, a lower level may lose the candidate or lead to a quick promotion; a higher level may set expectations the evidence does not yet support. That example shows the form only; use the evidence.
5. What to check before deciding: criteria nobody tested, with a question that would settle each.

Keep it under 600 words, apart from the table. A good write-up lets someone who missed the debrief see why the level was chosen.

Use only the interview evidence and the level definitions. Do not use age, graduation year, years since graduating or any protected characteristic as evidence, and count years of experience only where the level definitions name it. Do not use the candidate's current or past pay as a reason for a level, because many places restrict asking about it and it can carry old pay gaps forward. If interviewers disagree about the same criterion, show both views rather than averaging them. If the level definitions are missing or too vague to test against, say so at the top and list what they would need to say.

Variables

{{candidate_name}}
The candidate's name or a reference such as "Candidate B".
{{role}}
The role being hired for.
{{decision_maker}}
Who decides the level, for example "the hiring manager, with the head of department".
{{level_definitions}}
Your written definitions for the levels in question, criterion by criterion.
{{interview_evidence}}
The panel's feedback and scorecards, labeled by interviewer role.
{{levels_considered}}
The levels in question, for example "Senior or Staff".
{{pay_ranges}}
The pay range for each level being considered.
{{team_levels}}
How many people sit at each level in this role, for example "4 at Senior, 1 at Staff".

Why it works

  • Format. A criterion-by-criterion table puts the evidence for each level side by side, so the panel's disagreement becomes something the decision maker can read.
  • What to avoid. Age, graduation year, protected characteristics and past pay are ruled out as evidence, so the level rests on what the candidate showed in interviews.
  • If unsure. Conflicting views are shown rather than averaged, and vague level definitions are flagged before any recommendation is made.

How to use it

  1. Press Fill in blanks and type your details into each blank. What you type stays in this browser.
  2. Copy the filled prompt into the AI tool you already use, such as ChatGPT, Claude or Gemini, or open it in the Grader or the Prompt Tester here.
  3. Before you rely on it, try it on three or four real examples. The Prompt Tester runs them all at once and shows which ones go wrong.
Use it from code

Get this prompt as JSON from the free REST API, with your own text in the blanks, or let an AI assistant fetch it through the MCP server. No key.

curl -s https://thinkhard.si/api/v1/prompts/hiring-level-decision

With the MCP server added to Claude, ChatGPT, Cursor or VS Code, call get_prompt with {"prompt": "hiring-level-decision"}, or just ask for it by name.