Beginner 10 minutes Works with: ChatGPT, Claude, Gemini

How to Use AI to Write Job Descriptions

Quick Answer

Give AI the role title, what the person will do daily, what success looks like, and who your company is. Use the prompt below to generate a job description that’s specific enough to filter the wrong candidates out and compelling enough to draw the right ones in.

What This Workflow Helps You Do

  • Write clear, specific job descriptions faster
  • Avoid generic responsibility lists that attract unqualified applicants
  • Communicate company culture and expectations honestly
  • Create screening questions that reveal real candidate fit

Tools You Can Use

ChatGPT: fast and effective for most standard job descriptions.

Claude: better for senior roles or descriptions that need a more nuanced tone.

Gemini: useful if you’re posting across Google-integrated job platforms.

What You Need Before Starting

  • Role title
  • What the person will do day-to-day (3–5 main responsibilities)
  • What success looks like in 30, 60, and 90 days
  • Required vs. nice-to-have skills
  • Team structure and who they report to
  • Salary range and location/remote status
  • Company description (2–3 sentences)

Step-by-Step Process

Step 1: Build your input brief

Role: [job title]
Team: [who they work with and report to]
Day-to-day responsibilities: [list 4–6 main tasks]
Success in 90 days looks like: [specific outcomes, not vague goals]
Must-have qualifications: [non-negotiables]
Nice-to-have: [preferred but not required]
Company: [2–3 sentences about what you do and your culture]
Salary: [$X–$Y or "competitive + equity"]
Location: [remote / hybrid / on-site / city]

Step 2: Generate the job description

Write a job description using this brief:
[paste your brief]

Structure:
1. Role overview (2–3 sentences — who you are, what this person does, why it matters)
2. What you'll do (5–6 specific responsibilities — real day-to-day tasks, not generic)
3. What success looks like (2–3 concrete outcomes in the first 90 days)
4. What we're looking for (required and nice-to-have, separated clearly)
5. What we offer (compensation, benefits, culture — honest and specific)
6. About us (2–3 sentences)

Tone: [professional / friendly / direct]
Length: 400–500 words

Avoid: "fast-paced environment", "self-starter", "rockstar", "ninja", 
"wear many hats", "competitive salary" without numbers

Step 3: Add a culture signal

Generic job posts look identical. Add one honest line about what your company is actually like to work at:

Write 3 one-sentence culture signals for this job description.
They should tell a candidate what it's actually like to work here — 
the good AND a realistic expectation. Honest and specific, not marketing copy.

Company reality: [describe your actual culture in 2–3 sentences]

Step 4: Generate screening questions

Based on this job description:
[paste the final description]

Write 5 screening questions to include in the application.
Each question should reveal whether a candidate genuinely fits the role —
not just whether they've done the job title before.

Include 1 question that tests for the most critical skill.
Include 1 question that reveals how they handle [a specific challenge of this role].

Example Output

Role Overview: We’re a 12-person SaaS company looking for a Customer Success Manager to own relationships with our top 50 enterprise accounts. You’ll be the person clients call when something’s wrong, and the reason they renew when contracts come up.

What you’ll do:

  • Run onboarding for new enterprise accounts (first 60 days)
  • Conduct monthly check-in calls and quarterly business reviews
  • Identify expansion opportunities and work with sales on upsells
  • Monitor product usage and flag at-risk accounts before they churn
  • Document client feedback and relay product requests to the team

What success looks like in 90 days:

  • All 50 accounts have had a QBR
  • NPS from your accounts is above 40
  • 3 at-risk accounts have been moved to healthy status

Output will vary based on your role and company inputs.

Editing Checklist

  • Role overview is specific to this role, not copy-paste from another JD
  • Responsibilities describe real daily work, not just “own X”
  • Success metrics are concrete and measurable
  • Required skills are truly required (not a wishlist)
  • Salary range is included, hiding it reduces qualified applicants
  • No filler phrases (“fast-paced”, “rockstar”, “competitive salary”)

Common Mistakes

A wishlist for requirements. Listing 15 requirements filters out qualified candidates and signals poor scoping. Pick the 4–5 that are actually non-negotiable.

No success metrics. “Manage relationships” is vague. “Ensure 90% renewal rate on your account portfolio” is not.

Hiding the salary. Candidates increasingly filter jobs by salary range. Not including it means more unqualified conversations.

Generic culture descriptions. “We’re a team of passionate people who love what we do” says nothing. Be honest about pace, expectations, and what type of person succeeds here.

Continue learning

Explore related guides, tools, workflows, and prompts that help you go deeper into this topic.

Browse more AI workflows

Explore step-by-step guides for writing, research, marketing, content, and more.

Frequently Asked Questions

What makes a job description attract better applicants?

Specificity. Vague job descriptions attract anyone and filter nobody. Be specific about what the role does day-to-day, what success looks like in 90 days, and what kind of person thrives in this environment. Avoid generic lists of responsibilities.

How long should a job description be?

400–600 words is the sweet spot. Long enough to give candidates a real picture of the role, short enough to hold attention. AI tends to write long, trim anything that sounds like filler or that every job post says.

Should I use AI to screen applications too?

You can use AI to help write screening questions and evaluation criteria after the job post is written. Paste the job description and ask AI to generate 5 screening questions that reveal whether a candidate actually fits the role.

Last updated: