diff --git a/data/evidence/testset_fable.jsonl b/data/evidence/testset_fable.jsonl index e69de29b..a04de247 100644 --- a/data/evidence/testset_fable.jsonl +++ b/data/evidence/testset_fable.jsonl @@ -0,0 +1,20 @@ +{"hard_skills": ["full lifecycle recruiting", "candidate screening", "sourcing strategies", "vacancy announcement development", "relationship building"], "tools": ["Hirebridge ATS", "applicant tracking systems"], "methods": ["structured interviews", "proactive outreach", "social media recruiting"], "responsibilities": ["job requisition management", "candidate onboarding", "job posting maintenance", "candidate pipeline development"], "qualifications": ["2+ years full lifecycle recruiting experience", "bachelor's degree in human resources", "experience with federal government contracting recruitment"], "soft_skills": ["attention to detail", "communication skills", "flexibility", "composure"], "seniority": "junior", "job_id": "4MvxcXrkBQyplaHUAAAAAA=="} +{"hard_skills": ["technical recruiting", "candidate sourcing", "resume review", "interviewing", "negotiation", "communication"], "tools": ["applicant tracking system (ATS)"], "methods": ["phone screening", "face-to-face interviewing", "job fair attendance"], "responsibilities": ["candidate attraction", "candidate evaluation", "talent pipeline management", "hiring manager coordination", "offer presentation", "job description revision"], "qualifications": ["7+ years technical recruiting experience", "aerospace and defense industry experience", "dod secret clearance (desired)", "bachelor's degree or equivalent experience"], "soft_skills": ["strong communication", "negotiation skills", "independent work", "teamwork", "prioritization"], "seniority": "senior", "job_id": "PJzBWGtVOLQSCkJ_AAAAAA=="} +{"hard_skills": ["job requisition creation", "candidate assessment", "internet job posting", "recruiting"], "tools": [], "methods": [], "responsibilities": ["hiring manager consultation", "job description review", "candidate assessment advice"], "qualifications": ["bachelor's degree in human resources", "6+ years corporate recruiting experience"], "soft_skills": [], "seniority": "mid", "job_id": "zDyXJD_M-bo_BP9aAAAAAA=="} +{"hard_skills": ["recruitment", "legal compliance", "candidate sourcing", "resume review", "interview coordination", "offer negotiation", "performance management"], "tools": ["applicant tracking system", "LawCruit", "LinkedIn", "social media", "job boards"], "methods": ["inclusion and opportunity strategies", "market mapping", "direct outreach", "onboarding program", "orientation program"], "responsibilities": ["recruitment process management", "candidate experience delivery", "hiring strategy development", "agency spending tracking", "relationship building", "seamless onboarding", "due diligence support", "team supervision"], "qualifications": ["bachelor’s degree", "5+ years recruiting experience", "J.D. with 3+ years legal practice"], "soft_skills": ["communication skills", "attention to detail", "client service", "responsiveness", "interpersonal skills", "diplomacy", "adaptability"], "seniority": "mid", "job_id": "4kBNRyZTKQfUx78HAAAAAA=="} +{"hard_skills": ["full lifecycle recruiting", "candidate screening", "sourcing strategies", "vacancy announcement development", "relationship building"], "tools": ["Hirebridge ATS", "applicant tracking systems"], "methods": ["structured interviews", "proactive outreach", "social media recruiting"], "responsibilities": ["job requisition management", "candidate onboarding", "job posting maintenance", "candidate pipeline development"], "qualifications": ["2+ years full lifecycle recruiting experience", "bachelor's degree in human resources", "experience with federal government contracting recruitment"], "soft_skills": ["attention to detail", "communication skills", "flexibility", "composure"], "seniority": "junior", "job_id": "WVs1HkNCPHh2X4v5AAAAAA=="} +{"hard_skills": ["full-cycle recruiting", "sourcing", "resume screening", "candidate interviewing", "workforce planning", "offer negotiation"], "tools": ["applicant tracking system (ATS)", "ADP Workforce Now", "LinkedIn"], "methods": ["employer branding", "recruiting events"], "responsibilities": ["requisition intake", "hiring manager partnership", "candidate pipeline development", "interview coordination", "candidate communication", "pipeline management", "compliance adherence"], "qualifications": ["3+ years full-cycle recruiting experience", "bachelor's degree in human resources"], "soft_skills": ["communication skills", "interpersonal skills", "organizational skills", "time management skills", "attention to detail", "professionalism"], "seniority": "mid", "job_id": "e2WtN4CNdf0WSj_6AAAAAA=="} +{"hard_skills": ["full-cycle recruiting", "sourcing", "market research", "candidate assessment"], "tools": ["HARRI (applicant tracking system)"], "methods": [], "responsibilities": ["recruitment for leadership roles", "talent pipeline development", "hiring manager collaboration", "candidate experience delivery", "requisition management", "hiring metrics tracking"], "qualifications": ["2–4 years recruiting experience", "bachelor’s degree preferred"], "soft_skills": ["communication skills", "organizational skills", "time management skills"], "seniority": "mid", "job_id": "mnixAbGfHEd_dpy6AAAAAA=="} +{"hard_skills": ["recruitment coordination", "candidate sourcing", "interview scheduling", "data management", "compliance", "documentation"], "tools": ["applicant tracking system (ATS)"], "methods": ["full-cycle recruitment"], "responsibilities": ["job posting", "candidate communication", "hiring process coordination", "recruitment data maintenance"], "qualifications": ["bachelor’s degree in human resources", "2–4 years of experience in recruitment", "experience with applicant tracking systems"], "soft_skills": ["organizational skills", "attention to detail", "communication"], "seniority": "mid", "job_id": "wrYT3khPfsTFqqBxAAAAAA=="} +{"hard_skills": ["candidate sourcing", "talent acquisition", "data analysis", "market trend analysis", "recruiting strategy"], "tools": ["Microsoft Excel", "Microsoft Word", "Microsoft PowerPoint", "Microsoft Outlook", "Microsoft Teams", "Zoom", "Greenhouse", "LinkedIn / LinkedIn Recruiter"], "methods": ["benchmarking", "process improvement", "best practices"], "responsibilities": ["candidate communication", "interview scheduling", "candidate evaluation", "recruiting data tracking", "reporting and analysis", "employer branding"], "qualifications": ["bachelor's degree", "master's degree", "2 years related experience"], "soft_skills": ["written communication", "oral communication", "teamwork", "detail orientation", "problem solving", "empathy", "time management", "confidentiality"], "seniority": "junior", "job_id": "dh_CdH6Moz7V_tOcAAAAAA=="} +{"hard_skills": ["full-cycle recruiting", "sourcing", "screening", "interviewing", "job description creation", "candidate pipeline management", "recruiting metrics tracking"], "tools": ["Greenhouse", "LinkedIn / LinkedIn Recruiter", "Clearancejobs.com"], "methods": ["structured interviewing", "applicant tracking system (ATS)", "compliance standards", "process improvement"], "responsibilities": ["intake process management", "hiring manager partnership", "job posting", "candidate sourcing", "interview scheduling", "offer support", "recruiting record maintenance", "stakeholder communication"], "qualifications": ["2+ years recruiting experience", "experience with applicant tracking system (ATS)"], "soft_skills": ["communication skills", "interpersonal skills", "organizational skills", "relationship building"], "seniority": "mid", "job_id": "sEB-s90ujQDnAtL8AAAAAA=="} +{"hard_skills": ["candidate sourcing", "full lifecycle recruiting", "talent pipeline development", "job offer negotiation", "recruiting data analysis"], "tools": ["applicant tracking system (ATS)", "Google Suite", "Microsoft Excel"], "methods": [], "responsibilities": ["hiring manager partnership", "candidate screening", "talent brand marketing", "sourcing diverse candidates", "status update provision"], "qualifications": ["bachelor’s degree", "3+ years recruiting experience", "2+ years ATS experience"], "soft_skills": ["influencing", "verbal communication", "written communication", "organization skills", "time management"], "seniority": "mid", "job_id": "Pqf227A1zhx5E9JVAAAAAA=="} +{"hard_skills": ["full lifecycle recruiting", "sourcing", "screening", "interviewing", "talent pipeline management", "compliance", "workforce planning"], "tools": ["iCIMS", "LinkedIn Recruiter", "Indeed", "Cleared Jobs", "Microsoft Excel", "Microsoft Word"], "methods": ["social media sourcing", "internet sourcing", "progressive sourcing methods", "performance metrics optimization"], "responsibilities": ["recruiting strategy execution", "candidate pipeline building", "talent pool development", "offer negotiation", "hiring manager consultation", "compliance adherence"], "qualifications": ["bachelor's degree or equivalent", "5+ years recruiting experience", "experience with government contracts", "experience sourcing cleared personnel", "experience with government proposals"], "soft_skills": [], "seniority": "senior", "job_id": "Fh_u0B_8Mj3Sei9OAAAAAA=="} +{"hard_skills": ["sourcing", "full life cycle recruiting", "technical requirements gathering", "candidate evaluation", "data analysis"], "tools": ["LinkedIn Recruiter", "GitHub", "Stack Overflow", "applicant tracking system (ATS)"], "methods": ["proactive sourcing", "pipeline development"], "responsibilities": ["sourcing strategy execution", "talent pipeline maintenance", "hiring manager partnership", "candidate screening", "metrics tracking", "data accuracy maintenance"], "qualifications": ["5+ years recruiting experience"], "soft_skills": ["communication skills", "analytical skills"], "seniority": "senior", "job_id": "34biBMOuXaVCgy4bAAAAAA=="} +{"hard_skills": ["technical recruiting", "sourcing", "candidate pipeline development", "structured assessment", "workforce planning"], "tools": ["applicant tracking system (ATS)"], "methods": ["evidence-based evaluation", "full-cycle recruiting", "creative outreach", "referral programs"], "responsibilities": ["hiring manager partnership", "success profile definition", "interview loop design", "candidate experience creation", "recruiting systems improvement"], "qualifications": ["demonstrated success recruiting technical talent", "experience hiring engineers", "experience hiring product managers", "experience hiring designers", "experience hiring researchers"], "soft_skills": [], "seniority": "mid", "job_id": "TEaWx7-c_VoadWWCAAAAAA=="} +{"hard_skills": ["candidate generation", "market research", "talent mapping", "resume screening", "phone screening", "recruiting metrics tracking"], "tools": ["LinkedIn", "applicant tracking system (ATS)", "job boards"], "methods": ["diversity initiatives", "global recruiting practices"], "responsibilities": ["hiring needs understanding", "candidate pipeline building", "ats record maintenance", "interview coordination", "recruitment event support"], "qualifications": ["1+ years recruiting experience", "bachelor's degree"], "soft_skills": ["communication skills", "stakeholder management"], "seniority": "mid", "job_id": "4rr0nqfO5LBt_RhvAAAAAA=="} +{"hard_skills": ["full-cycle recruiting", "candidate sourcing", "resume screening", "interviewing", "offer negotiation", "employment law", "data analysis"], "tools": ["LinkedIn Recruiter", "Wellfound", "applicant tracking system (ATS)", "Google Suite"], "methods": ["structured interviews", "proactive sourcing", "talent pipeline management"], "responsibilities": ["job brief creation", "candidate screening", "interview coordination", "process refinement", "data tracking", "relationship building", "candidate communication"], "qualifications": ["5+ years full-cycle recruiting experience"], "soft_skills": ["written communication", "verbal communication", "organization", "discretion", "professionalism"], "seniority": "senior", "job_id": "GhsPPrf49qxJoB4AAAAAAA=="} +{"hard_skills": ["talent acquisition", "candidate screening", "interview preparation", "compensation negotiation", "sourcing", "relationship building"], "tools": ["Microsoft Outlook", "Microsoft Teams", "Microsoft Word", "Microsoft Excel", "job boards", "recruiting software"], "methods": ["full life-cycle recruiting"], "responsibilities": ["talent pipeline development", "candidate matching", "recruiting strategy development", "candidate evaluation"], "qualifications": ["college degree", "experience with high-volume recruiting", "sourcing experience", "interviewing/screening experience", "candidate management experience"], "soft_skills": ["self-motivation", "communication skills", "organizational skills", "enthusiasm", "persuasion"], "seniority": "mid", "job_id": "_nfE8xkyCEjwnz2MAAAAAA=="} +{"hard_skills": ["candidate sourcing", "passive candidate generation", "talent assessment"], "tools": [], "methods": [], "responsibilities": ["pipeline development", "market insights leveraging", "engagement of top talent"], "qualifications": [], "soft_skills": [], "seniority": "mid", "job_id": "scdn6Wf-ohb3uW9dAAAAAA=="} +{"hard_skills": ["full-cycle recruitment", "behavioral interviewing", "skillset assessment", "sourcing", "candidate screening", "market analysis", "talent recommendations", "job description writing"], "tools": ["Lever", "Dice", "LinkedIn Recruiter"], "methods": ["principle based management (PBM)", "bottom-up empowerment"], "responsibilities": ["hiring manager partnership", "application review", "interview scheduling", "candidate communication", "recruitment strategy execution", "documentation maintenance"], "qualifications": ["5+ years recruitment experience"], "soft_skills": ["prioritization", "communication", "judgment", "flexibility", "entrepreneurialism"], "seniority": "mid", "job_id": "-KvDWHXJiCgCYKzjAAAAAA=="} +{"hard_skills": ["full-cycle recruitment", "sourcing", "screening", "interviewing", "offer negotiation", "onboarding", "staffing strategy", "risk management", "compliance", "data analysis", "workforce planning"], "tools": ["applicant tracking system (ATS)", "CRM", "Symplr CTM", "iCIMS", "Sense", "Jira"], "methods": ["performance management", "coaching", "continuous improvement", "contract staffing", "talent pipeline development"], "responsibilities": ["team leadership", "recruitment oversight", "staffing solutions delivery", "labor market analysis", "candidate closing", "proposal support", "process improvement", "vendor management"], "qualifications": ["bachelor’s degree", "5+ years healthcare recruiting experience", "3+ years managing recruiting teams"], "soft_skills": ["leadership", "communication", "stakeholder management"], "seniority": "lead", "job_id": "Qb3-ggfQVWeA6hRxAAAAAA=="} diff --git a/pipeline/extract_local.py b/pipeline/extract_local.py index c4beb5ae..79b6e1e4 100644 --- a/pipeline/extract_local.py +++ b/pipeline/extract_local.py @@ -39,6 +39,11 @@ ARRAY_FIELDS = { "responsibilities": (10, 80), "qualifications": (8, 80), "soft_skills": (8, 40), } SENIORITY = {"junior", "mid", "senior", "lead", "n/a"} +# gemma3 copies these prefixes into qualifications although they are skills, +# not formal requirements (testset 2026-07-07: 1/20 records) -> drop them there +NOT_A_QUALIFICATION = ("strong ", "understanding of", "knowledge of", "familiarity with") +ACRONYM_TOOLS = {"ats": "applicant tracking system (ATS)", "crm": "CRM", + "hris": "HRIS", "erp": "ERP", "cad": "CAD", "pos": "POS"} def validate(rec): @@ -65,7 +70,11 @@ def validate(rec): for item in val: if not isinstance(item, str): continue - item = re.sub(r"\s+", " ", item).strip(" .,;:-") + item = re.sub(r"\s+", " ", item.replace("�", "'")).strip(" .,;:-") + if field == "qualifications" and item.lower().startswith(NOT_A_QUALIFICATION): + continue + if field == "tools": + item = ACRONYM_TOOLS.get(item.lower(), item) if not (2 <= len(item) <= max_len): continue key = item.lower()