Prompts for hiring your first employees

By the Chat Judge team · Updated · 4 min read · Links go to prompts in our library; creators are credited on each prompt.

In short: Hiring has a lot of repeated writing and a judgement you should not delegate. These prompts handle the writing and the structure: whether the role is justified, a job description built on outcomes, outreach with the pay range stated, an interview scorecard everyone uses, a fair offer and a 30-60-90 day plan. The decision about the person stays with you.

Decide before you write the job post

Decide if this is the right next hire challenges the decision rather than polishing it. You describe the problem the role would solve, the work you currently do yourself, the budget and the alternatives you have considered, such as a contractor, an agency, a tool or simply waiting. It asks whether your evidence supports hiring now, what would make it clearly yes, and which single outcome would justify the cost. Use it before you spend time on a job description you may not need.

The hiring prompts in this guide were adapted by us from the MIT-licensed gtm-skills project by LeadMagic and rewritten for founders hiring their first employees. Sources and licence terms are on the credits page.

The job description and the post

Job description that attracts builders structures the post around why the role exists now, the outcomes the person will own in their first months, what a great week looks like, a short list of must-haves and an honest note about the stage. It asks for results rather than a list of duties, and it includes the pay range, because leaving pay out costs you the candidates who will not ask.

Once the full description exists, LinkedIn hiring post compresses it into a short blurb that points people to the full post. Write the long version first. The short one is a summary, and summaries of a vague description stay vague.

Reaching people who are not looking

Outreach to a passive candidate writes three messages: a first note under 100 words that names something specific about their work, states the outcome the role owns and the pay range, and makes a small ask; a follow-up a few days later; and a third that closes the loop politely. The specific detail has to be real and checked by you. A message that compliments the wrong project is worse than no message.

Be careful with personal data. Use a placeholder for the candidate's name while you draft, and check the data settings of whatever tool you paste into.

Interviews, the offer and the first 90 days

Structured interview scorecard builds the thing that keeps hiring fair: five or six competencies tied to the outcomes, two questions every candidate is asked, descriptions of strong, medium and weak answers, and a rule that interviewers score before they discuss. Consistency is the point. An unstructured chat favours whoever interviews best, not whoever will do the job.

Interview plan and work sample designs the whole process inside a time budget and includes one realistic work sample based on a real problem you face, with instructions, a time limit and scoring. Keep the work sample short. Candidates are giving you time they are not paid for.

Prepare an offer and anticipate the counter assembles the total package, explains each part in plain language and predicts the requests you are most likely to hear, with a response for each. Decide your limits before the conversation. 30-60-90 day plan then gives the new person what they should learn, deliver and who they should meet in each month, plus a first-week checklist. Check that the plan matches the outcomes you hired them for, since those are the terms you both agreed to.

Run the loop in order

The prompts form a sequence, and skipping ahead creates rework. First decide whether the role should exist. Then write the description from the outcomes you just defined, and only then the short post and the outreach. Design the scorecard and the work sample before the first interview, not after the first candidate, or you will judge people against different standards. The offer comes after you have chosen, and the 30-60-90 plan should be ready before their first morning.

Two safeguards are worth writing down. Every interviewer scores before the debrief, using the same questions, and you keep brief notes of the evidence rather than a gut feel. And anything a prompt drafts about a real person, such as an outreach message or an offer, gets a human read for accuracy and tone before it is sent. The prompts are adapted from MIT-licensed open source and rewritten for early teams; they are a structure, not a hiring policy, so check the employment and pay rules that apply where you hire.

How to get good results from any of these prompts

  1. Fill in every placeholder. Anything in [BRACKETS] is a fact only you know. Leave one empty and the model will guess.
  2. Tell it what not to invent. Add "use only the facts above, and flag anything missing" if the prompt does not already say it.
  3. Ask for one version first. Fix the direction, then ask for variations.
  4. Check every name, number and promise before anything leaves your hands.
  5. Edit it into your own voice. Read it aloud once; if it does not sound like you, change it.

Prompts in this guide

Questions people ask

Can AI pick the best candidate?

No. Use it for consistent questions, scorecards and write-ups, then decide on the evidence from interviews yourself.

How do I keep interviews fair?

Ask every candidate the same questions, use the same scale, score before the group discussion and write down evidence. Check the employment and pay-transparency rules where you hire.

Should I share the pay range in outreach?

The outreach prompt includes it, because it saves time for both sides. Where you hire may also have rules about pay transparency.

Keep exploring

More guides

Written by the Chat Judge team from prompts in our library. We do not run model benchmarks, and nothing here is a claim about results you will get. Spot a problem? Tell us or submit a better prompt.