Shortlisting CVs responsibly
Listen to this lesson
Nigerian voice narration
Fictional training case. 240 applications arrived for one vacancy and the published criteria require a first shortlist of 20 within 48 hours. HR risks both delay and inconsistent scoring if each reviewer judges CVs differently, so an anonymised, rubric-based AI pass is used to create a defensible first cut before the panel meets. Work through the example, identify the evidence needed and practise the method with cleared or synthetic inputs. Any additional figures in the explanation are separate practice examples, not claimed rows in a downloadable file.
Worked example
Convert the published vacancy criteria into a numbered rubric (e.g. minimum qualification, years of experience, required certification) worth fixed points each. Strip names, photos, states of origin, and any identifying detail from the applicant summaries before pasting them in batches. Ask AI to score each summary strictly against the rubric and show its reasoning per criterion, not an overall gut feeling. Export the scores into a spreadsheet, sort descending, and flag anyone within 2 points of the cut-off line as 'borderline' for the panel to review by hand rather than auto-excluding them.
The method
Anonymise-then-rubric-score: remove identity markers first, then force the model to score against your fixed criteria list only, requesting a reason per line. This can help expose bias and vague 'overall impression' scoring that cannot be defended later.
Worked practice and common mistakes
Use fictional CV summaries for practice; identity removal alone does not authorise uploading live applications. Define the scoring scale and tie handling before reviewing candidates. Ask for source evidence per criterion and mark missing evidence as unknown instead of assuming a score. The two-point borderline window is an example, not a universal recruitment rule. Follow the approved panel process.
Review before using the result
Spot-check 10 scored CVs manually against the rubric. If AI scores disagree with your manual read on more than one or two, tighten the rubric wording and rerun that batch before the panel sits.
Your workplace practice
Use the recommended free assistant with this lesson's fictional example. Ask for the output described in the worked example. Keep a copy of the input and draft. Compare the result with the verification steps above, record any correction, and retain placeholders where evidence is missing. The expected result is a checked draft for human review, not an automatically approved decision. Course examples and practice responses are fictional; amounts, dates, rules and thresholds are not official guidance or supplier quotations.
Why this tool: Can organise fictional candidate evidence against a rubric; a human panel checks the evidence and makes decisions.
Hit a free limit? Switch to ChatGPT or Gemini and continue — the prompt works in any assistant.