# Halve candidate screening time: the AI talent acquisition build for in-house and agency recruiters Source: https://getstartedwithai.net/tutorials/ai-talent-acquisition-build Site: Get Started With AI Tagline: Build AI automations that actually run your business. Audience: solo professionals, freelancers, consultants and AI implementers in training Contact: hello@getstartedwithai.co.uk This Markdown companion is generated from the prerendered page for AI crawlers. # Halve candidate screening time: the AI talent acquisition build for in-house and agency recruiters In-house talent acquisition runs more repeatable manual work per role than almost any other knowledge function in a UK SMB. The build below moves CV triage to a Custom GPT scored against a written rubric, produces a ranked shortlist with a one-paragraph reason per candidate and drafts personalised outreach for the top names. Final shortlist decisions stay with the hiring manager and the TA partner; the time saved sits on the reading-and-sorting layer that absorbs most of the day. Built in 90 minutes; runs at £35 a month plus a fortnight of output review for accuracy and bias mitigation. ## What you need - A ChatGPT Plus or Claude Pro account at £16 a month. - A Zapier paid plan and your existing ATS (Workday, Greenhouse, Lever, SmartRecruiters, Pinpoint or Teamtailor). - The job spec template and 30 anonymised CVs (15 strong, 15 borderline) for calibration. - A privacy notice update disclosing AI use plus a UK GDPR Article 22 note keeping a human in the loop on the final decision. - 90 minutes uninterrupted, then 30 minutes per role on every shortlist after. ## Step 1: Write the TA rubric (20 minutes) In-house TA cares about culture and trajectory more than agency desks do. Build that into the rubric or the GPT will optimise for the wrong thing. - Hard requirements: right to work, qualification, named tech or skill stack. - Strong indicators: 4 to 6 markers the hiring manager looks for in a strong hire (employer type, project type, years in seat). - Trajectory signals: career pattern, gap explanations, lateral moves vs progressions. - Disqualifiers: explicit reasons to reject regardless of fit. ## Step 2: Build the screening Custom GPT (25 minutes) 1. Open ChatGPT, create a Custom GPT named TA Screening Rubric. 2. Upload the rubric, the spec, the brand voice notes for the outreach line and 30 calibration CVs with the hiring manager scores. 3. System prompt: you are screening UK CVs against an in-house TA rubric. UK English. Output: candidate ID, score, one-paragraph reason, suggested outreach line in the brand voice. Anonymise the candidate name and contact detail; reference by ID only. 4. Add the rule: keep a human in the loop on the final shortlist; never recommend an offer. 5. Test against the 30 calibration CVs. Hiring manager agreement should land around 85 to 90 percent in the first run. ## Step 3: Wire the ATS pipeline (20 minutes) 1. In Zapier, trigger on a new candidate against a requisition in the ATS. 2. Action 1: anonymise the CV. Action 2: send to the GPT with the rubric inserted. Action 3: write the score, the reason and the outreach line to the candidate record. 3. Action 4: tag the requisition with the count of shortlisted candidates ready for the hiring manager. ## Step 4: The bias-audit fortnight (15 minutes setup) The fortnight of output review is what gets the build to production accuracy and where the bias mitigations land for in-house compliance. 1. Add a daily Sheet log of every AI-suggested rejection with the rubric reason, the candidate ID and the named reviewer. 2. The TA partner reads every rejection across the fortnight. Where the AI rejected wrongly, log the reason and update the rubric. 3. After the fortnight, accuracy lands at around 95 percent and the bias audit log goes into the firm AI register. ## Step 5: First production roles Run the build on three live roles. Compare the AI shortlist against the manual shortlist. The third role is where the build stops surprising the TA partner and starts saving real time. ## What you ship A Custom GPT with the TA rubric, a Zap into the ATS, an anonymised review queue, a bias audit Sheet, a privacy notice update and an AI register entry. Cuts CV screening time roughly in half across the TA function. ## What to build next Plug in the recruitment firm build for any agency partners on the same desk, the AI for finance team build for the headcount budget side, or move into the WIRE Method to package the TA overlay across requisitions and offices. ## Related builds More tutorials in Industries. Each one runs in 60 to 90 minutes and ships a finished asset. ### Reclaim 8 hours a week as an accountant: the AI build for practice work 30 to 60 minutes. Solo Pro / Implementer. [Open tutorial](/tutorials/ai-for-accountants-practice-build) ### Cut 12 hours a week of legal admin: the AI build for solicitors and small firms 30 to 60 minutes. Solo Pro / Implementer. [Open tutorial](/tutorials/ai-for-lawyers-firm-build) ### Replace 6 hours a week of session admin: the AI build for coaches and therapists 60 to 90 minutes. Solo Pro. [Open tutorial](/tutorials/ai-for-coaches-therapists-build) ### Reclaim a day a week as a property agent: the AI build for letting and sales 60 to 90 minutes. Solo Pro / Implementer. [Open tutorial](/tutorials/ai-for-real-estate-agents-build) ## Questions people ask before they start ### What does talent acquisition AI cover? AI tools used in the recruitment process across CV triage, candidate sourcing, outreach personalisation, interview note-taking and offer-stage analytics. The build above covers the first three; sourcing tools (LinkedIn Recruiter AI, hireEZ) sit alongside rather than inside it. ### Will AI take recruiter jobs? No. AI replaces the high-volume, low-judgement parts of the TA role including initial CV triage and outreach drafting, not the candidate conversation or the hiring manager partnership. Firms that adopt AI run more roles per TA per quarter rather than reduce headcount. ### Is AI in recruitment legal in the UK? Yes, with disclosure requirements. The candidate has a right to know AI was used in the decision; the build ships with a one-paragraph disclosure for the candidate-facing privacy notice. UK GDPR Article 22 limits fully-automated decisions; the build keeps a human in the loop on the final shortlist for that reason. ### Does this work with Workday, Greenhouse, Lever or Pinpoint? Yes across the major UK in-house ATS platforms. Greenhouse, Lever and Pinpoint have Zapier or API patterns that fit the screening trigger. Workday needs a Make HTTP module pattern because of the slightly heavier auth, but the same Custom GPT layer applies. ### How long until the build pays back? Around 6 weeks for an in-house TA function running 5 plus open requisitions a month. The build cost is one half-day plus £35 a month; the saving is half the screening time per role from week three.