Purpose of AI in CometHire

CometHire uses AI to support structured analysis of application documents and evidence-based candidate review workflows. Our AI features are designed to assist recruiters during screening and evaluation processes while keeping humans fully in control of all hiring decisions.

AI-Supported Workflows

CometHire uses large language models (LLMs) to analyze job descriptions and application documents in order to support structured candidate evaluation.

AI-supported features may help users:

• generate and structure evaluation criteria
• highlight relevant information within application documents
• provide criterion-level evidence assessments (Limited Evidence, Partial Evidence, High Evidence)
• suggest interview or reference check questions
• summarize candidate qualifications and experiences
• generate structured candidate reports

All AI-supported outputs are intended to support human review and interpretation. Recruiters remain responsible for defining evaluation criteria, reviewing candidate information, and making final decisions.

Human Oversight

CometHire is intentionally designed as a human-in-the-loop system. The software does not automatically reject, rank or select candidates based on AI evaluation without human involvement. The software does not generate candidate rankings or suitability scores.

For each recruiter-defined requirement, CometHire provides an AI-supported assessment of the amount of evidence found in the application documents. Evidence levels are presented as qualitative categories ("Limited Evidence", "Partial Evidence", or "High Evidence") and are linked to supporting information found in the application. These evidence assessments are intended to support human review and do not constitute candidate rankings, suitability scores or hiring recommendations.

Users can review, adjust, ignore or manually overwrite AI-supported outputs and recommendations at any time.

Anonymization & Bias Reduction

To support more skills-focused evaluations, CometHire offers anonymized screening workflows that can reduce visible identifying information such as names, contact details, age, gender and profile images during early-stage candidate review.