GenAI for Recruitment
15 production-ready prompts. 8 end-to-end lifecycle workflows. A 30-day team adoption roadmap. Bringing efficiency, speed, and intelligence to every stage of hiring.
A note before you begin
AI will not replace you. Someone using AI will. This guide is built so you become the latter — confidently, ethically, and on your own terms.
The workshop you just attended was deliberately fast-paced. This resource is the slower companion. Treat it as a working document. Mark it up. Argue with it. Adapt the prompts to your business. The point is not to use AI the way we showed you. The point is to make AI a default in how you work.
How to use this guide
- Start with what's broken. Pick one recruiter task that drains your week — JD writing, screening, follow-ups — and apply the matching workflow first.
- Copy the prompts. Every prompt below is designed to paste into ChatGPT, Claude, or Gemini and adapt in under a minute.
- Run the 30-day plan. Don't try to transform your function in a week. The roadmap gives you a sequenced rollout that compounds.
- Come back to the principles. Tools change. The mindset doesn't.
Document 1 of 3
The Prompt Pack
15 prompts · 5 categories. Production-ready GenAI prompts that recruiters can copy, paste, and adapt across the full hiring lifecycle.
How to use this pack
- Copy the prompt directly into ChatGPT, Claude, Gemini, or your preferred GenAI tool.
- Replace the
[square brackets]with your specific role, context, or candidate details. - Iterate — if the first output isn't perfect, refine with follow-ups like 'make it more concise' or 'rewrite for senior-level audience'.
- Always review AI outputs before sending to candidates or hiring managers. AI assists, humans decide.
- Never paste confidential candidate data (PII, salaries, internal notes) into public AI tools.
01 JD Drafting & Optimization
Create compelling, structured, and bias-free job descriptions in minutes.
1.1
Generate a Complete JD from a Role Brief
Use case: When a hiring manager gives you a vague brief and you need a full job description fast.
Act as a senior recruiter for a [industry] company. Create a detailed job description for the role of [job title] based on the following inputs: years of experience [X], key responsibilities [list], must-have skills [list], nice-to-have skills [list]. Structure the JD with these sections: About the Role, Key Responsibilities, Required Qualifications, Preferred Qualifications, What Success Looks Like in 90 Days, and Why Join Us. Use inclusive, gender-neutral language. Keep tone professional but warm. Output in clean markdown.💡 Pro tip: Always paste your company's culture statement and EVP into the same chat first — the AI will weave it naturally into the 'Why Join Us' section.
1.2
Rewrite an Existing JD for Better Engagement
Use case: When an existing JD is generic, jargon-heavy, or failing to attract quality applicants.
Below is a job description that is underperforming on our careers page. Rewrite it to be more candidate-centric, compelling, and inclusive. Specifically: shorten responsibility lists to maximum 6 bullets, replace corporate jargon with plain English, lead with impact rather than tasks, and add a clear 'Day in the Life' section. Highlight growth opportunities and learning. Original JD: [paste JD here].💡 Pro tip: Ask the AI to also produce a one-line 'hook' you can use as a LinkedIn job post headline.
1.3
Audit a JD for Bias and Inclusivity
Use case: Before publishing any JD externally to ensure you're not unintentionally narrowing your talent pool.
Act as a DEI consultant. Review the following job description and flag any language that may be: gendered or coded masculine/feminine, ageist, ableist, exclusionary to non-traditional career paths, or that signals 'culture fit' bias. For each flagged item, suggest a neutral alternative. Also identify any 'must-haves' that could be reframed as 'nice-to-haves' to broaden the candidate pool. JD: [paste JD here].💡 Pro tip: Run this audit even on JDs you've used for years — many roles still contain subtle bias from older templates.
02 Rapid Resume Screening
Move from a stack of CVs to a ranked shortlist in a fraction of the time.
2.1
Score a Single Resume Against a JD
Use case: When you have one candidate's resume and need a fast, structured fit assessment.
Act as an experienced recruiter. Compare the resume below against the job description and produce: (1) an overall fit score from 1 to 10 with one-sentence reasoning, (2) a table mapping each must-have skill in the JD to specific evidence in the resume, marking each as Strong / Partial / Missing, (3) the candidate's top 3 strengths for this role, (4) the top 3 gaps or risks, and (5) three specific interview questions to probe the gaps. JD: [paste JD]. Resume: [paste resume].💡 Pro tip: Save the output as a recruiter brief and attach it to the candidate record in your ATS for hiring manager review.
2.2
Bulk Rank a Shortlist of Resumes
Use case: When you have 5–10 resumes for the same role and need to prioritize who to call first.
I will paste 5 resumes below for the role of [job title]. Rank them from best to worst fit against this JD: [paste JD]. For each candidate, provide: rank, fit score out of 10, one-sentence verdict, top strength, biggest gap, and a recommended next step (call now / phone screen / reject with feedback). Output as a clean markdown table sorted by rank. Resumes: [paste resumes separated by '---'].💡 Pro tip: Limit to 5–10 resumes per prompt — quality of ranking degrades with very large batches. Run multiple rounds for bigger pipelines.
2.3
Extract a Structured Profile from a Messy Resume
Use case: When a resume is poorly formatted and you need clean data for your ATS or a hiring manager email.
Parse the following resume and extract the data into a structured format with these fields: Full Name, Current Title, Current Company, Total Years of Experience, Relevant Years of Experience for [target role], Top 5 Technical Skills, Top 3 Soft Skills, Notable Achievements (quantified where possible), Education, Certifications, and Career Progression Summary in 2 sentences. Output as clean JSON and as a recruiter-friendly summary. Resume: [paste resume].💡 Pro tip: The JSON output can be pasted directly into a Google Sheet for instant pipeline tracking.
03 Structured Candidate Evaluation
Turn unstructured interview impressions into rigorous, comparable evaluations.
3.1
Generate a Role-Specific Interview Scorecard
Use case: Before any panel interview to ensure consistent, criteria-based evaluation across all interviewers.
Act as a Head of Talent. Create an interview scorecard for the role of [job title] at a [company stage / industry] company. Include 5 to 7 evaluation competencies relevant to this role, with each competency containing: a clear definition, what 'great' looks like, what 'average' looks like, what 'concerning' looks like, and 2 sample probing questions. Use a 1-4 rating scale (1=Below Bar, 2=Mixed, 3=Solid, 4=Exceptional). Format as a clean table I can paste into Notion or Google Docs.💡 Pro tip: Share the scorecard with all panel members 24 hours before the interview so each interviewer can claim 2 competencies to focus on.
3.2
Compare Two Finalist Candidates
Use case: When you have two strong finalists and the hiring manager is stuck on a decision.
I have two finalist candidates for the role of [job title]. Compare them across these dimensions: technical skills, leadership signals, growth trajectory, cultural add (not fit), risk factors, salary expectations vs. budget, and likelihood of acceptance. Provide a side-by-side comparison table, then a final recommendation with reasoning, and finally one tie-breaker question I should ask each candidate before deciding. Candidate A: [paste]. Candidate B: [paste].💡 Pro tip: Ask the AI to also produce a 5-minute verbal summary you can deliver to the hiring manager — saves you a meeting prep block.
3.3
Synthesize Panel Feedback into a Hiring Recommendation
Use case: After all interviews are done to produce a clear, defensible decision document.
Below are interview notes from [N] panel members for candidate [name] applying for [role]. Synthesize them into: (1) a one-paragraph executive summary, (2) consensus strengths cited by 2+ interviewers, (3) consensus concerns cited by 2+ interviewers, (4) outlier opinions worth investigating, (5) overall hiring recommendation (Strong Hire / Hire / No Hire / Strong No Hire) with reasoning, and (6) any reference-check questions worth asking before extending an offer. Notes: [paste all notes].💡 Pro tip: If interviewer notes disagree sharply, ask the AI to flag potential bias signals — e.g., one interviewer rating dramatically lower on subjective dimensions.
04 Automated Bias Reduction
Build fairness checks directly into your daily recruiting workflow.
4.1
Anonymize a Resume for Blind Screening
Use case: When you want hiring managers to evaluate candidates on substance only, before names are visible.
Anonymize the following resume by removing or masking: full name (replace with 'Candidate A'), gender indicators, photographs, age, date of birth, marital status, nationality, religious affiliations, names of schools/universities (replace with 'Tier 1 University' / 'Tier 2 University' / 'Tier 3 University' based on global rankings), specific company names where they could signal prestige bias (replace with 'Top 5 [Industry] Firm' or similar). Preserve all skills, achievements, responsibilities, and quantified results exactly as written. Resume: [paste resume].💡 Pro tip: Run blind screening for the first round only — names and full context can return at the shortlist stage where bias risk is lower.
4.2
Audit Interview Questions for Bias
Use case: Before any interview to catch questions that could disadvantage certain candidates.
Below is a list of interview questions planned for the role of [job title]. Review each question and flag any that: are not job-relevant, could disadvantage candidates from non-traditional backgrounds, assume specific life experiences (e.g., parenting, military service), use idioms that may not translate across cultures, or test 'culture fit' rather than 'culture add'. For each flagged question, suggest a fairer alternative that tests the same underlying competency. Questions: [paste list].💡 Pro tip: Re-run this audit annually — what felt neutral two years ago may not feel neutral today.
4.3
Rewrite Rejection Feedback to Be Constructive and Bias-Free
Use case: When you want to give every rejected candidate genuinely useful feedback without legal risk.
I'm rejecting a candidate for [role]. Here are the panel's notes: [paste]. Draft a rejection email that: opens with genuine appreciation, gives 2-3 specific, behaviour-based reasons (never personality, identity, or 'fit' based), offers 1-2 concrete suggestions for how they could strengthen their candidacy in future, leaves the door open for relevant future roles, and closes warmly. Tone: respectful, human, not corporate. Keep under 200 words. Avoid any language that could be construed as discriminatory.💡 Pro tip: Have your legal/HR team review the first 3 rejection templates the AI produces — once approved, you can scale safely.
05 Advanced Sourcing Strategy
Reach the candidates your competitors are missing.
5.1
Generate Boolean Search Strings for LinkedIn
Use case: When sourcing for niche roles where standard keyword searches return too much noise.
Act as a sourcing specialist. Generate 3 Boolean search strings for finding [job title] candidates on LinkedIn with the following requirements: [list must-haves]. The strings should target: (1) the obvious candidate pool, (2) adjacent talent pools (people doing similar work under different titles), and (3) underutilized talent pools (e.g., career returners, candidates from adjacent industries). For each string, explain who it will surface and what trade-offs it makes. Use proper Boolean syntax with AND, OR, NOT, and parentheses.💡 Pro tip: Save winning Boolean strings in a team-shared doc tagged by role family — they're a sourcing asset that compounds in value over time.
5.2
Map Adjacent Talent Pools
Use case: When your obvious candidate pool is exhausted and you need to think laterally.
I'm hiring for the role of [job title] and the traditional talent pool is dry. Suggest 8-10 adjacent talent pools where I might find capable candidates with transferable skills. For each, provide: the source pool (industry, function, or background), the transferable skills they bring, the gaps they would have, the upskilling effort required, the typical compensation delta, and one example of a search query to find them. Be creative but practical — focus on pools that are realistic to convert, not just theoretically interesting.💡 Pro tip: Bring this list to the hiring manager early — adjacent-pool hires often need slightly modified scorecards and onboarding plans.
5.3
Craft a Personalized Outreach Message
Use case: When reaching out to a passive candidate whose attention you only get one shot at capturing.
Draft a personalized LinkedIn InMail to a passive candidate with the following profile: [paste their headline, current role, and 2-3 specifics from their profile]. The role I'm pitching: [job title] at [company], with this hook: [list 2-3 things that make this role genuinely interesting]. Requirements for the message: under 100 words, opens with a specific reference to their work (not flattery), states the role and the hook in one tight sentence, ends with a low-friction ask (15-min chat, not 'apply now'), and feels human — not templated. Avoid the words 'rockstar', 'ninja', 'opportunity', and 'leverage'.💡 Pro tip: If response rates stay below 15%, the issue is usually not your message — it's the role itself or the wrong candidate fit. Re-evaluate before iterating further.
Document 2 of 3
Lifecycle Templates
8 end-to-end AI-powered workflows. Step-by-step recruitment workflows that show exactly where AI fits, what each step produces, and how to scale across your team.
The 8 Workflows at a Glance
1Role Briefing & JD ComparisonAlign hiring manager and recruiter on role expectations before sourcing begins
2Sourcing Strategy PlanningBuild a multi-channel sourcing plan with talent pool maps and outreach assets
3Evaluation & Shortlist ReportsConvert resume screening into ranked shortlists with structured rationale
4Custom Interview Question BanksGenerate role-specific behavioural and technical question sets per competency
5Engagement & Pre-BoardingMaintain candidate momentum from offer through Day 1 with personalized touchpoints
6Offer Drafts & KPI DashboardsProduce offer letters and track recruitment KPIs in a single workflow
7Rejection & Feedback LoopsClose the loop with rejected candidates while protecting employer brand
8Post-Hire 90-Day ReviewConnect quality-of-hire signals back into the sourcing and evaluation playbook
01
Role Briefing & JD Comparison
Align hiring manager and recruiter on what 'success in this role' truly means before a single candidate is sourced.
When to use: At kickoff for any new role, especially when the JD has been recycled, the hiring manager is hiring for this role for the first time, or two stakeholders disagree on what the role actually needs.
| # | Step | Action & AI Assist | Output |
|---|---|---|---|
| 1 | Capture the brief | 30-min hiring manager intake meeting. Record (with consent) or take detailed notes covering: outcomes expected in 12 months, must-have skills, deal-breakers, comp range, and similar people the manager admires. | Raw intake notes |
| 2 | Synthesize with AI | Paste raw notes into GenAI: 'Synthesize this hiring manager intake into a structured role brief with sections: Role Outcome, Must-Have Skills, Nice-to-Haves, Red Flags, Comp Band, Interview Loop.' | Structured role brief |
| 3 | Draft the JD | Use the brief to generate a candidate-facing JD with the AI. Apply Prompt 1.1 from the Prompt Pack. | Draft JD v1 |
| 4 | Bias audit | Run JD through Prompt 1.3 to flag biased or exclusionary language and rewrite for inclusivity. | Bias-cleaned JD |
| 5 | Compare to similar roles | Ask AI: 'Compare this JD to typical [role] postings at top companies. What's missing or unusual?' | Gap analysis |
| 6 | Manager sign-off | Send the brief + JD to the hiring manager with three specific questions to confirm alignment. Lock the version. | Approved JD v1.0 |
02
Sourcing Strategy Planning
Build a multi-channel sourcing plan with talent pool maps and ready-to-send outreach for every channel.
When to use: Within 24 hours of finalizing the JD. Especially valuable for niche, senior, or high-volume roles where one channel alone won't fill the funnel.
| # | Step | Action & AI Assist | Output |
|---|---|---|---|
| 1 | Map talent pools | Use Prompt 5.2 to identify primary, adjacent, and underutilized talent pools for the role. | Pool map (10 pools) |
| 2 | Build Boolean strings | Use Prompt 5.1 to generate 3 Boolean strings tuned to LinkedIn and one to Google X-Ray. | 4 search strings |
| 3 | Channel allocation | Ask AI: 'Given these talent pools and a 2-week sourcing window, suggest a channel mix and target candidate count per channel.' Allocate effort across LinkedIn / Job boards / Referrals / Communities. | Channel plan |
| 4 | Outreach asset library | Use Prompt 5.3 to draft 3 InMail variants: cold, warm (mutual connection), and re-engagement (prior applicants). Add an email variant and a referral ask. | 5 outreach templates |
| 5 | Set sourcing cadence | Define daily and weekly targets per channel. Ask AI to draft a sourcing tracker template. | Sourcing tracker |
| 6 | Launch & measure | Begin outreach. Track response rate, meeting booked rate, and shortlist conversion by channel. Iterate weekly. | Live pipeline |
03
Evaluation & Shortlist Reports
Turn a stack of resumes and interview notes into a ranked, defensible shortlist with clear rationale per candidate.
When to use: Twice in every search — once after first-round screening (long list to short list), and again before the hiring manager debrief (short list to finalists).
| # | Step | Action & AI Assist | Output |
|---|---|---|---|
| 1 | Standardize input | Collect all resumes in one folder. Convert non-text resumes (PDF scans) to text using OCR or AI vision. | Clean resume set |
| 2 | Bulk rank | Use Prompt 2.2 to rank batches of 5–10 resumes against the JD with fit scores and rationale. | Ranked long list |
| 3 | Anonymize finalists | For top 8–10, use Prompt 4.1 to anonymize for blind hiring manager review. | Anonymized profiles |
| 4 | Generate recruiter brief | For top 5, use Prompt 2.1 to produce a single-page recruiter brief per candidate (strengths, gaps, suggested probes). | 5 recruiter briefs |
| 5 | Synthesize into shortlist report | Ask AI: 'Combine these 5 briefs into one shortlist report for hiring manager review, with a comparison matrix and a recommended top 3.' | Shortlist report |
| 6 | Manager review meeting | Walk hiring manager through the matrix in 15 minutes. Capture their reactions for the next batch's prompt tuning. | Approved finalists |
04
Custom Interview Question Banks
Generate competency-mapped, role-specific interview questions in minutes — and keep them improving over time.
When to use: Before kicking off the interview loop for any role, especially when the loop will involve multiple panel members who need to coordinate.
| # | Step | Action & AI Assist | Output |
|---|---|---|---|
| 1 | Pull competencies from JD | Ask AI: 'Extract 5–7 core competencies from this JD and define each in one sentence.' | Competency list |
| 2 | Generate question bank | For each competency, ask AI to produce 5 behavioural questions, 3 technical/case questions, and 2 follow-up probes. | 60–80 questions |
| 3 | Build the scorecard | Use Prompt 3.1 to generate a panel scorecard with 1–4 rating definitions per competency. | Scorecard v1 |
| 4 | Audit for bias | Use Prompt 4.2 to flag any questions that risk disadvantaging certain candidates. | Cleaned bank |
| 5 | Allocate to panel | Assign 1–2 competencies per panel member so coverage is complete without redundancy. Document who covers what. | Panel coverage map |
| 6 | Capture & improve | After each loop, ask panel which questions surfaced the best signal. Feed those back into the bank for the next role. | Improving question bank |
05
Engagement & Pre-Boarding
Keep candidate momentum high from first touch through Day 1 with personalized, well-timed communications.
When to use: Throughout the entire pipeline. Critical between offer-accept and Day 1 — the highest-risk window for ghosting and offer renege.
| # | Step | Action & AI Assist | Output |
|---|---|---|---|
| 1 | Map every touchpoint | List every moment a candidate hears from you: application ack, screen invite, post-screen update, interview confirm, post-interview, offer, pre-boarding (weekly), Day 0 welcome. | Touchpoint map |
| 2 | Generate templates | For each touchpoint, ask AI to draft a warm, on-brand template with merge fields. Tone: human, specific, no corporate filler. | 8–10 templates |
| 3 | Personalize at scale | For high-priority candidates, paste their LinkedIn or resume into AI and ask for a personalized opening line per outreach. | Personalized variants |
| 4 | Build the pre-board sequence | Between offer-accept and Day 1, schedule weekly 'we're excited' touches: team intro video, equipment confirmation, first-day logistics, manager video greeting. | Pre-board calendar |
| 5 | Detect ghost signals | If a candidate goes quiet >72 hours, ask AI to draft a low-pressure check-in that doesn't feel desperate. | Recovery message |
| 6 | Day 1 handoff | Send a final pre-Day-1 brief to the hiring manager and onboarding buddy: candidate's interests, motivations, and one personal detail. | Day 1 handoff doc |
06
Offer Drafts & KPI Dashboards
Produce consistent, compliant offer letters and track recruitment health metrics from the same workflow.
When to use: Offer drafting: every time a verbal offer is accepted in principle. KPI dashboard: weekly review cadence, with deeper analysis monthly.
| # | Step | Action & AI Assist | Output |
|---|---|---|---|
| 1 | Capture offer terms | Confirm: base, variable, equity, joining bonus, benefits, start date, location, reporting line, notice handling. | Offer terms checklist |
| 2 | Generate offer letter | Ask AI: 'Draft an offer letter using these terms and our standard template structure. Flag any clauses that need legal review.' | Draft offer letter |
| 3 | Legal & comp review | Route the draft for legal sign-off and comp committee approval as per company policy. | Approved offer |
| 4 | Send & track | Send via your offer management tool. Log: sent date, response date, accept/decline, reasons if declined. | Offer log entry |
| 5 | Weekly KPI dashboard | Each Friday, paste the week's pipeline data into AI: 'Generate a recruitment KPI summary covering open roles, time-to-hire, source effectiveness, offer accept rate, and pipeline health. Flag any metric that has worsened week-on-week.' | Weekly KPI brief |
| 6 | Monthly trend review | Roll up 4 weekly briefs into a monthly trend report. Identify 1 systemic improvement to implement next month. | Monthly review + 1 action |
07
Rejection & Feedback Loops
Close the loop with rejected candidates in a way that protects employer brand and creates future pipeline.
When to use: For every candidate who is rejected at any stage. Especially important for candidates who completed interviews — they invested time and deserve closure.
| # | Step | Action & AI Assist | Output |
|---|---|---|---|
| 1 | Tier the rejection | Classify into: pre-screen reject (form letter), post-screen reject (short personal email), post-interview reject (detailed personal email with feedback). | Rejection tier |
| 2 | Draft tier-3 rejection | For interview rejections, use Prompt 4.3 to draft constructive, bias-free feedback based on panel notes. | Personalized rejection |
| 3 | Manager review (if needed) | For senior roles or borderline rejections, route the rejection draft for hiring manager sight before sending. | Approved rejection |
| 4 | Send & log | Send within 5 business days of decision. Log the candidate as 'silver medalist' if they would be reconsidered for a different role. | ATS update |
| 5 | Talent community add | For silver medalists, send a follow-up offering to add them to the talent community for future relevant roles. Ask their permission first. | Community opt-in |
| 6 | Quarterly re-engagement | Every quarter, ask AI to scan the talent community for matches against currently open roles. Re-engage with personalized notes. | Re-engagement list |
08
Post-Hire 90-Day Review
Connect quality-of-hire signals back into your sourcing, screening, and evaluation playbook for compounding improvement.
When to use: 90 days after every new hire's start date. This is the workflow that turns recruitment from a transactional function into a learning system.
| # | Step | Action & AI Assist | Output |
|---|---|---|---|
| 1 | Schedule the check-in | At Day 60, set up three 15-min surveys: new hire, hiring manager, onboarding buddy. | 3 scheduled surveys |
| 2 | Run surveys | Standard questions: ramp-up speed, expectation match, support quality, predicted retention, biggest strength, biggest gap. | 9 data points / hire |
| 3 | Synthesize with AI | Paste survey responses into AI: 'Synthesize these three perspectives into a quality-of-hire summary. Flag any disconnects between the three views.' | QoH summary |
| 4 | Trace back to source | Tag the hire with: source channel, recruiter, interview panel, top scorecard competencies, comp bracket. Look for patterns. | Tagged hire record |
| 5 | Quarterly pattern analysis | Each quarter, ask AI to analyze 90-day reviews across all hires and identify: best-performing sources, scorecard items most predictive of success, weakest interview signals. | Quarterly playbook update |
| 6 | Update playbook | Refine: which channels to prioritize, which scorecard items to weight higher, which interview questions to retire. Communicate changes to the team. | v2 playbook |
How to roll these out
Don't try to operationalize all 8 workflows at once. Recommended sequence over 60–90 days:
- Weeks 1–2: Workflows 1 & 4 (Role Briefing + Interview Question Banks). Highest-leverage upstream artifacts.
- Weeks 3–4: Workflows 2 & 3 (Sourcing Strategy + Evaluation/Shortlists). Compound on weeks 1–2.
- Weeks 5–6: Workflows 5 & 6 (Engagement + Offers/KPIs). Tighten the conversion engine.
- Weeks 7–8: Workflows 7 & 8 (Rejection Loops + Post-Hire Review). Build the long-term learning system.
Document 3 of 3
30-Day Adoption Roadmap
A day-by-day plan to embed AI into your recruitment team's daily workflow — with structured ownership, check-ins, success metrics, and ongoing upskilling.
The Five Pillars of Adoption
- Structured workflow changes — concrete, documented changes to how the team operates day-to-day.
- Clear ownership definitions — every workflow has a named owner accountable for outcomes.
- Concrete success measures — metrics that everyone agrees define 'this is working'.
- Weekly team check-in cadence — a fixed rhythm that keeps adoption visible and prevents drift.
- Ongoing AI upskilling paths — every team member knows what they're learning next, and why.
Week 1 · Days 1–7
Foundation
Install the basics — tools, prompts, mindset. The week where habits begin.
| Day | Focus | Activities | Owner |
|---|---|---|---|
| 1 | Kickoff & Tool Setup | 60-min team kickoff. Walk through this roadmap. Confirm everyone has access to a paid GenAI tool (ChatGPT Plus, Claude Pro, or Gemini Advanced). | Team Lead |
| 2 | Prompt Engineering Basics | 90-min hands-on workshop on the Role + Context + Format prompt structure. Each recruiter writes 3 weak prompts and 3 structured versions. | Team Lead |
| 3 | Pick Your 3 Use Cases | Each recruiter picks 3 prompts from the Prompt Pack to use this week on real work. Submit picks to the team channel. | Each Recruiter |
| 4 | First Live Use | Use AI on at least one real task today. Save the prompt + the output. Note time saved vs. manual approach. | Each Recruiter |
| 5 | Show & Tell | 30-min team session. Each person demos one AI output they created this week. No judgement — just sharing. | Team Lead |
| 6 | Self-Reflection | (Light day) Capture: what worked, what felt awkward, where AI gave bad output and why. | Each Recruiter |
| 7 | Week 1 Check-In | 45-min review meeting. Run the Week 1 check-in checklist. Lock in Week 2 commitments. | Team Lead |
Week 1 check-in
- Every team member has working access to their AI tool.
- Every team member has used AI on at least one live recruitment task.
- We have a shared channel or doc where prompts are being saved.
- Each person has named the 3 specific use cases they will run in Week 2.
- No one feels behind — if anyone is stuck, we have a buddy plan in place.
Week 2 · Days 8–14
Implementation
Take AI from experiment to embedded — apply it to the full lifecycle of one live role each.
| Day | Focus | Activities | Owner |
|---|---|---|---|
| 8 | Pick Your Pilot Role | Each recruiter picks one live, currently open role to run end-to-end with AI. This becomes their Week 2 case study. | Each Recruiter |
| 9 | JD with AI | Use Prompt Pack Category 1 to draft or rewrite the JD for the pilot role. Run the bias audit. Get hiring manager sign-off. | Each Recruiter |
| 10 | Sourcing with AI | Use Prompt Pack Category 5 to build Boolean strings, map adjacent talent pools, and craft outreach messages for the pilot role. | Each Recruiter |
| 11 | Screening with AI | Use Prompt Pack Category 2 to score and rank incoming resumes. Compare AI ranking with your gut — note disagreements. | Each Recruiter |
| 12 | Interview Kit with AI | Use Prompt Pack Category 3 to generate the scorecard and question bank for the pilot role. Share with the panel 24 hours before interviews. | Each Recruiter |
| 13 | Mid-Sprint Demo | 60-min team session. Each recruiter presents their pilot role progress: what AI helped with, what it didn't, how much time was saved. | Team Lead |
| 14 | Week 2 Check-In | 45-min review. Run the Week 2 check-in checklist. Pick the templates that will become team standards. | Team Lead |
Week 2 check-in
- Every recruiter has used AI across at least four lifecycle stages on their pilot role.
- We have specific time-saved data points to share — not just 'it felt faster'.
- Hiring managers on the pilot roles have given feedback on the AI-generated artifacts.
- We've identified at least 3 prompts that worked exceptionally well for our context.
- We've identified at least 1 prompt that consistently failed and needs adaptation.
Week 3 · Days 15–21
Integration
Move from individual experiments to team standards. Build the playbook everyone uses.
| Day | Focus | Activities | Owner |
|---|---|---|---|
| 15 | Build the Prompt Library | Consolidate the prompts that worked into a single team-wide prompt library (Notion, Confluence, or Google Doc). Tag by lifecycle stage. | Team Lead + 1 Recruiter |
| 16 | Standardize Templates | Pick the AI-generated templates that will become team standards: JD format, scorecard, recruiter brief, rejection email, offer letter. | Team Lead |
| 17 | Implement Workflow #1 | Roll out Workflow 1 (Role Briefing) as a team standard. Every new role kickoff this week uses it. | Each Recruiter |
| 18 | Implement Workflow #3 | Roll out Workflow 3 (Evaluation & Shortlist Reports) as a team standard. Every shortlist this week is delivered using this format. | Each Recruiter |
| 19 | Implement Workflow #4 | Roll out Workflow 4 (Custom Interview Question Banks) as a team standard. Every interview loop launching this week uses an AI-generated scorecard. | Each Recruiter |
| 20 | Bias & Quality Audit | 60-min team session. Audit AI-generated outputs from the past two weeks for bias, accuracy, and quality. Document red flags and refinements needed. | Team Lead + DEI Partner |
| 21 | Week 3 Check-In | 45-min review. Run the Week 3 check-in checklist. Confirm playbook v1.0 is live. | Team Lead |
Week 3 check-in
- The team prompt library is published and everyone knows where to find it.
- Three lifecycle workflows are now team standards — not individual experiments.
- Every new role kicked off this week followed the AI playbook.
- The bias audit has been completed and any red flags have a written remediation plan.
- Hiring managers across multiple roles are aware that we're using AI and have not raised concerns.
Week 4 · Days 22–30
Optimization & Scale
Measure what worked, refine what didn't, and plan how AI fluency keeps growing beyond Day 30.
| Day | Focus | Activities | Owner |
|---|---|---|---|
| 22 | Measure 30-Day Impact | Compile metrics: time-to-shortlist, time-to-hire, screening time saved, hiring manager satisfaction, candidate NPS where available. | Team Lead |
| 23 | Refine Top 5 Prompts | Take the 5 most-used prompts and refine them based on 3 weeks of real use. Update the prompt library. | Each Recruiter |
| 24 | Retire the Failures | Identify prompts and workflows that haven't worked. Document why, retire them, and capture lessons. | Team Lead |
| 25 | Hiring Manager Feedback | Run a 15-min survey or interview with 3+ hiring managers who experienced AI-augmented recruitment. Capture quotes and suggestions. | Team Lead |
| 26 | Plan the Upskilling Path | Based on what worked, define each recruiter's next 30-day learning focus: e.g., advanced prompting, building a Custom GPT, learning a no-code AI workflow tool. | Team Lead + Each Recruiter |
| 27 | Stakeholder Showcase | 30-min showcase to TA leadership and key hiring managers. Share metrics, stories, lessons. Position the team as AI-fluent. | Team Lead |
| 28 | Plan the Next 30 Days | Define Days 31–60 commitments: 3 new workflows to standardize, 2 new tools to evaluate, 1 cross-team partnership to launch. | Team Lead |
| 29 | Document Playbook v1.0 | Combine the prompt library, standardized templates, workflows, and metrics into a single Recruitment AI Playbook v1.0 document. | Team Lead + 1 Recruiter |
| 30 | Day 30 Retrospective | 60-min team retrospective. Celebrate wins, name struggles honestly, commit to Days 31–60. Schedule next monthly check-in. | Whole Team |
Day 30 check-in
- We have hard numbers showing the impact of AI on time-to-hire and recruiter productivity.
- Every recruiter knows what they are working on for Days 31–60 and why.
- TA leadership has seen the showcase and is bought in to continued investment.
- The Recruitment AI Playbook v1.0 is published and lives in our shared workspace.
- We have a monthly check-in scheduled for the next 6 months to keep adoption alive.
Common Pitfalls — and How to Avoid Them
1. Treating Day 30 as the End
AI fluency erodes if it isn't reinforced. Teams that stop intentional upskilling at Day 30 see prompt quality and adoption drop within 90 days. Fix: schedule monthly AI retrospectives indefinitely, even when things feel fine.
2. Letting AI Replace Judgement
AI's most dangerous failure mode is confident wrongness. Teams that stop reviewing outputs eventually ship a biased JD, a wrong shortlist, or a tone-deaf rejection. Fix: keep human review on every AI artifact that touches a candidate or hiring manager. Always.
3. Hoarding Prompts
When recruiters keep their best prompts to themselves, the team's collective AI capability stagnates. Fix: build sharing into the weekly cadence — every check-in includes 'one prompt I'm sharing this week'.
4. Pasting Confidential Data into Public AI
Salaries, internal performance notes, candidate PII, and confidential offer terms must never be pasted into a free-tier consumer AI. Fix: agree on what data is OK and what is off-limits before Week 1 kicks off, and put it in writing.
5. Skipping the Stakeholder Showcase
Without visible business impact at Day 30, leadership won't fund continued investment, and the program will quietly die. Fix: schedule the Day 27 showcase before Week 1 begins, so it's a fixed deadline that forces you to gather real metrics.
Final word
Thirty days is enough time to install new habits, but not enough time to make them permanent. The teams that win with AI in recruitment are the ones that treat Day 30 as the moment they stop being beginners — not the moment they stop learning.
Used together, the Prompt Pack, Lifecycle Templates, and 30-Day Roadmap reliably produce: faster hiring, better shortlists, fairer evaluations, and a team that genuinely enjoys their job more — because the boring parts of recruitment are finally getting handled by the tool, and the human parts are getting more of their attention.
Want help rolling this out?
If your team wants a structured rollout with hands-on workshops and live coaching, we run end-to-end AI adoption programs.
Speak to Our Team →