Bias and Inclusion Scan — review job descriptions

Bias and Inclusion Scan API

What it does

Bias and Inclusion Scan reviews a job description for biased or exclusionary language. Send the text of a role description, and you get back a structured assessment with a winner, a rationale, and a comparisonMatrix broken into experience, softSkills, and technicalSkills.

Use it when you want to screen hiring copy before it goes live, compare two versions of a posting, or check whether your language favors one type of candidate over another. The response is designed for downstream review flows: winner gives you the result, while rationale explains the decision in plain terms.

This is useful for recruiting teams, ATS integrations, and internal HR tooling that need a fast consistency check on job ads. It helps you catch wording that may discourage qualified applicants and gives reviewers a concise breakdown of where the language is skewed.

Bias and Inclusion Scan is a simple way to turn subjective job-post review into a repeatable API step.

Quickstart

Compare the submitted text for bias and inclusion by sending it as JSON.

curl -X POST "http://localhost:8080/namastesumalya/scan-bias-and-inclusion" \
  -H "apy-token: $APY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"text":"We are looking for a rockstar developer who can work 80 hours a week."}'

What you'll get back

Returns a JSON object with winner and rationale string fields, plus an optional comparisonMatrix object. When present, comparisonMatrix contains experience, softSkills, and technicalSkills string fields.

{
  "winner": "B",
  "rationale": "Candidate B's alignment with modern cloud-native practices makes them a better fit for our future-state architecture.",
  "comparisonMatrix": {
    "experience": "A has deeper tenure (15y vs 8y), but B has more relevant modern stack experience.",
    "softSkills": "Both demonstrate strong leadership, but B shows better alignment with modern agile engineering cultures.",
    "technicalSkills": "A is legacy-enterprise focused; B is cloud-native and high-performance focused."
  }
}

About this endpoint

What it does

Scans the submitted job description text and returns a structured assessment with a winner, a rationale, and a comparisonMatrix object containing category-specific explanations.

Request Body

Parameter	Type	Mandatory	Description
text	String	Yes	The job description text to scan for bias and inclusivity.

Response

Returns a JSON object with three top-level fields: winner and rationale are strings, and comparisonMatrix is an object containing string fields for experience, softSkills, and technicalSkills.

Parameter	Type	Mandatory	Description
winner	String	No	The returned winner value.
rationale	String	No	The explanation for the result.
comparisonMatrix	Object	No	Category-level comparison details.
comparisonMatrix.experience	String	No	Comparison details for experience.
comparisonMatrix.softSkills	String	No	Comparison details for soft skills.
comparisonMatrix.technicalSkills	String	No	Comparison details for technical skills.

Body

Name

Type

Description

bodyREQUIRED

object

—

Quickstart

What you'll get back

About this endpoint

What it does

Request Body

Response

Body

Errors any endpoint can return