Tech & AIBeginnerPreview
ChatGPT for HR & Recruiting
A hands-on course for HR professionals, recruiters, and small-business owners who want AI to do the slow parts of hiring and people operations: writing job descriptions, screening resumes, building structured interview kits, drafting offer and onboarding materials, and producing HR templates. You leave with reusable prompts, fair-by-design workflows, and a document library you run for every requisition.
For recruiters, HR generalists, hiring managers, and small-business owners who run their own hiring and want AI to draft and screen faster without bias or legal risk.
Course content
Workbook & downloads
Put the course into practice — a printable workbook plus editable templates you can fill in and reuse.
Preview the workbook
This workbook turns the course into tools you use for every requisition: a job-description and bias-check kit, a fair screening scorecard, a structured interview kit with rubrics, candidate-communication templates, and a 30-60-90 onboarding plan. Work through one section per module, filling the worksheets and running the prompts in ChatGPT or Claude as you go. By the end you will have a complete, reusable hiring and HR workflow plus a prompt and document library you can run on autopilot — with the fairness and privacy guardrails built in.
Setting Up AI for People Work
Establish the fairness, legal, and privacy guardrails that govern every later exercise before you draft anything.
Worksheet: AI-in-Hiring Task Triage
List the HR and recruiting tasks you do, and sort each into one of three buckets. Anything in 'Do not automate' stays a human decision with AI only assisting around the edges. Revisit this whenever you adopt a new AI use.
- Safe for AI to draft (e.g. job descriptions, rejection emails)
- AI assists, I decide (e.g. resume screening, interview scoring)
- Do not automate (e.g. final hire, termination, discipline)
- Tool and tier I will use for HR data
- Who reviews and signs off on hiring decisions
Exercise: Audit One Job Post for Bias and Adverse Impact
Take an existing job description or a recent shortlist. Run the bias-check prompt on the wording, and informally check whether your recent hiring outcomes skew on any protected dimension. Note what you would change in the criteria, not just the candidates.
- Review this job description for gender-coded words, age signals, ableist requirements, and jargon; suggest neutral replacements and estimate reading level.
- List any requirement in this post that may not be genuinely job-related, with a one-line reason.
- Given these screening criteria, point out any that could act as a proxy for a protected characteristic.
Checklist: Fairness and Privacy Setup Checklist
- Confirmed which AI tools and tiers are approved for candidate and employee data
- Disabled model-training on inputs in any consumer AI account used for HR
- Adopted the rule: strip names and identifiers from resumes before screening
- Wrote down the principle that a human reviews every shortlist, score, and rejection
- Checked which hiring laws apply to you (EEOC, NYC LL144, EU AI Act, GDPR/CCPA/PIPEDA)
Writing Job Descriptions and Sourcing
Produce an accurate, inclusive job description, a clean bias check, and the sourcing outreach and search strings to fill your funnel.
Exercise: Draft a Job Description from a Real Role
Pick a role you are or will be hiring. Gather the real inputs first (team, first-year outcomes, truly required skills, salary range, location and work model, one honest reason to take the job). Run the job-description prompt, then run the bias check on the result and revise.
- Write a job description from these inputs, using outcome-focused bullets and inclusive second-person language; include the salary range.
- Now review that description for biased or exclusionary language and unnecessary requirements, with replacements.
- Trim the 'required' list to only what is genuinely necessary to do the job, and move the rest to 'nice to have'.
Worksheet: Job Description Brief
Complete every field before prompting. The quality of the job description depends entirely on this brief. Reuse this template for each new role.
- Role title and who it reports to
- Team and where it sits
- Three to five outcomes for the first year
- Genuinely required skills (must-haves only)
- Nice-to-have skills
- Salary range
- Location and work model (on-site, hybrid, remote)
- One honest reason this job is worth taking
Worksheet: Sourcing Outreach and Boolean Worksheet
Plan your sourcing before generating copy. Capture one real, verified personal detail per candidate and the search logic for the role, so AI personalizes truthfully and your search returns the right people.
- Role and the alternative job titles that also fit
- Must-have skills for the Boolean string
- Terms to exclude (NOT)
- Verified personal detail for the outreach message
- The one honest reason this role is interesting
- The soft ask (what you want them to do next)
Checklist: Pre-Posting Checklist
- Every requirement is genuinely job-related and defensible
- Bias check run and flagged language addressed
- Salary range and work model stated clearly
- Equal-opportunity statement and accommodation process included
- Reading level kept accessible (around an 8th-grade target)
Screening and Interviewing Candidates
Build a fair screening scorecard, a structured interview kit with rubrics, and an evidence-based candidate comparison.
Worksheet: Screening Scorecard
Translate the job description into five to eight job-related criteria, each with a clear definition, BEFORE reading any resume. This is the standard AI screens against. Score Strong, Partial, or Not evident, always backed by evidence from the resume.
- Criterion 1 and its definition
- Criterion 2 and its definition
- Criterion 3 and its definition
- Criterion 4 and its definition
- Criterion 5 and its definition
- Criteria 6-8 if needed
- What evidence counts as 'Strong' for each
- The cut line for the shortlist
Exercise: Screen a Resume Against the Scorecard
Mask a real resume's identifying details (label it Candidate A). Run the screening prompt against your scorecard. Read every rating's evidence quote yourself — never accept a score you cannot verify in the text. Note the criteria with missing evidence to probe in interview.
- Rate Candidate A against these defined criteria as Strong/Partial/Not evident, with one quote or fact as evidence for each.
- Do not consider name, age, school prestige, or employment gaps, and do not infer anything not written.
- List the criteria where evidence is missing so I know what to ask about.
Exercise: Generate a Structured Interview Kit
Choose four to five competencies for the role. Generate one behavioral question, two probes, and a 1-to-5 rubric per competency, then have AI sanity-check the questions against illegal categories. Use the same kit for every candidate.
- Create a structured interview kit: for each competency, one behavioral question, two follow-up probes, and a 1/3/5 rubric.
- Check every question against legal red lines (age, family, health, religion, origin) and flag any to remove.
- Add a one-line note on what a strong versus weak answer signals for each competency.
Worksheet: Candidate Comparison Grid
After interviews, record rubric scores and one line of evidence per competency for each finalist. Compare on the criteria, not on charisma. Note the open question each finalist must still resolve before a decision.
- Competencies listed down the side
- Candidate names across the top
- Rubric score (1-5) per competency per candidate
- One-line evidence behind each score
- Open question to resolve for each finalist (reference, final round)
- Who makes the final decision
Offers, Onboarding, and HR Documents
Close candidates well, onboard with a clear plan, and build a reusable HR document library with human and legal review.
Exercise: Draft an Offer and a Rejection
For one finalist, draft a warm written offer email using only confirmed terms, marking anything unconfirmed as [TO CONFIRM]. For a declined candidate, draft a prompt, respectful rejection that gives no protected-trait-linked reason. Edit both into your company's real voice.
- Draft a warm written offer email with title, start date, and salary; mark anything I leave blank as [TO CONFIRM] and invent nothing.
- Write a respectful, concise rejection for a candidate at the [stage], with no detailed individual feedback or protected-trait reason.
- Rewrite both to match this sample of our company voice that I paste.
Exercise: Build a 30-60-90 Onboarding Plan
For a new or upcoming hire, generate a 30-60-90 plan with learning goals, relationships, and deliverables per phase, plus a first-week checklist and three check-in points. Then have AI draft the supporting welcome email and FAQ.
- Create a 30-60-90 onboarding plan for this role and manager, with learning goals, key relationships, and deliverables per phase.
- Add a first-week checklist (accounts, tools, intro meetings) and 30/60/90-day check-in questions.
- Draft a welcome email, a first-day agenda, and a short role FAQ for the new hire.
Worksheet: Onboarding Roadmap
Lay out the new hire's first 90 days before generating copy, so the plan reflects the real role. Assign concrete deliverables to each phase rather than vague 'get up to speed' goals.
- First-week setup (accounts, tools, intro meetings)
- Days 1-30 learning goals and relationships
- Days 31-60 supported deliverables
- Days 61-90 owned outcomes
- Check-in dates and owner (manager, HR)
- Buddy or mentor and their role
Checklist: HR Document and Compliance Checklist
- Offer letter is factual; binding contract routed through legal review
- Every candidate who interviewed received a timely reply
- PIP and disciplinary documents reviewed by HR or legal before use
- AI was given real facts only — no fabricated incidents in any document
- Templates tailored to your jurisdiction, policies, and company voice
Your Action Plan
- Run the AI-in-hiring task triage and confirm which approved tool and tier you will use for candidate and employee data.
- Disable model-training on inputs and adopt the rule to mask resume identifiers before screening.
- Write a job-description brief for one open role and generate a draft, then run the bias check and revise.
- Build the sourcing outreach and Boolean strings for that role and save the working versions to your library.
- Turn the job description into a five-to-eight-criterion screening scorecard before reading any resume.
- Screen your current applicants against the scorecard with masked identities and a human review of every rating.
- Generate a structured interview kit with rubrics and sanity-check the questions against illegal categories.
- Interview with the kit, then build a candidate comparison grid from rubric-scored, evidence-based evaluations.
- Draft your offer and rejection templates and a 30-60-90 onboarding plan for the chosen hire.
- Start an HR document library (policies, PIP, survey, review templates) drafted with AI and finalized by a human reviewer.
Pairs well with
Courses members commonly take alongside this one.
Flagship CoursePreview
Freelance Business Foundations: Position, Price, Sell, and Deliver High-Value Services
Freelancing · Beginner · 16h
Self-pacedPreview
Client GrowthPreview
Freelance Client Acquisition: Outreach, Leads, Referrals, and Deal Flow
Freelancing · Beginner · 15h 30m
Self-pacedPreview
Sales SystemPreview
Freelance Sales & Proposals: Discovery Calls, Scoping, Objections, and Closing
Freelancing · Intermediate · 16h
Self-pacedPreview