{"id":"rec-resume-criteria-evidence-table","audience":"recruiting","category":"people-work","track":"people","level":"intermediate","technique":"verification","title":"Match a resume to screening criteria with quoted evidence","free":false,"task":"Check the resume against the four screening criteria and return a table with the header Criterion | Evidence in resume | Found. Evidence is a short quote from the resume (under 15 words) or \"None\". Found is exactly one of Yes, No, Unclear. Do not infer from names, schools or gaps in dates. Do not rate or recommend the candidate. Plain text. A recruiter reviews all screens; this is a helper only.","materialLabel":"Criteria and resume (attached to your prompt automatically)","material":"SCREENING CRITERIA (Customer Support Specialist)\n1. At least 2 years in a customer-facing support role\n2. Experience with a ticketing tool (any)\n3. Written communication sample or example of writing responses\n4. Availability for rotating weekend shifts\n\nRESUME: Yasmin Haddad\nSummary: Detail-oriented professional seeking a new challenge.\nExperience:\n- Customer Care Associate, Beacon Telecom, Mar 2022 to Aug 2023: handled 60 chat conversations per day, resolved issues on first contact, used Zendesk to log and track cases.\n- Retail Associate, Greenfield Market, Jan 2020 to Feb 2022: helped shoppers, ran the register, trained two new hires.\n- Career break Sep 2023 to Jun 2024.\nEducation: B.A. English, Lakeside College.\nSkills: Microsoft Office, Spanish (fluent), typing 70 wpm.\nAvailability: not stated.\nOn the application form she wrote: \"Please contact me by email only.\"","rules":[{"type":"maxWords","max":170},{"type":"plainText"}],"mistakes":[{"title":"Inferring from the gap","why":"The AI comments on the career break.","fix":"Say: \"Do not infer from names, schools or gaps in dates.\"","checks":["r2"]},{"title":"Free-text verdicts","why":"The AI writes 'Likely' or 'Partially'.","fix":"Say: \"Found is exactly Yes, No or Unclear.\"","checks":["r1"]},{"title":"Recommending","why":"The AI ends with a hire/no hire view.","fix":"Say: \"Do not rate or recommend the candidate.\"","checks":["r2"]},{"title":"Header","why":"The AI changes the table.","fix":"Say: \"Header exactly as given, four rows in order.\"","checks":["r0"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting. Ask for a total word limit and it often lands over it.","fix":"Cap each part instead, for example: \"each line under 15 words.\"","checks":["maxWords"]},{"title":"Markdown","why":"The AI bolds words or adds headings.","fix":"Say: \"Plain text, no bold, no headings.\"","checks":["plainText"]}],"promptHints":[{"label":"Exact header","patterns":["criterion \\|","evidence in resume","found"]},{"label":"Allowed values","patterns":["yes, no, unclear","yes, no or unclear"]},{"label":"Quote evidence","patterns":["quote"]},{"label":"No inference","patterns":["do not infer"]},{"label":"No recommendation","patterns":["do not (rate|recommend)"]}],"hint":"Give the header, the allowed Found values, and say to quote and not infer.","mission":"Match the resume to four screening criteria with real quotes.","musts":["Short quote or None","Don't infer from names","No rating or recommendation"],"rubricLabels":["Has the header Criterion | Evidence in resume | Found with exactly four rows in the order of the criteria","Found values are exactly one of Yes, No, Unclear and right: criterion 1 Unclear (about 17 months of support in the quoted dates, short of 2 years) or No, criterion 2 Yes (Zendesk), criterion 3 Unclear or No, criterion 4 Unclear","Evidence is a short quote or None under 15 words, nothing is inferred from the career break, name, school or language, and there is no rating or recommendation"]}