De-Identification & Screening

Before candidate applications are analyzed, CometHire applies an AI-supported de-identification step intended to reduce personally identifying information where possible. This may include attributes such as names, contact details, age, gender, and profile images.

The goal is to support more skills- and qualification-focused evaluations during early-stage screening processes.

Certain contextual information relevant to evaluating professional qualifications — such as previous employers, educational background, or project experience — may remain part of the analysis where necessary for the recruiting workflow.

Because the process is AI-supported, complete removal of all identifying or sensitive information cannot be guaranteed in every case.

Structured Candidate Evaluation

CometHire is designed to support more structured and comparable candidate evaluations by assessing candidates against predefined role-specific criteria.

Recruiters define and weight evaluation criteria based on the requirements of the role. The system analyzes application documents against these criteria and links evidence assessments to relevant information found in the application.

By combining structured evaluation criteria with evidence-based analysis, CometHire aims to reduce ambiguity and support more consistent evaluation workflows.

Ongoing Evaluation & Safeguards

We continuously evaluate and refine our AI-supported workflows to improve transparency, explainability, and consistency across different hiring scenarios.

Our approach combines selected third-party AI models with our own evaluation logic, prompting strategies, de-identification workflows, and product safeguards designed to support responsible and human-controlled recruiting processes.