About the Project

Inspiration

We are inspired by a simple but important observation. Many skilled candidates in India will continue to be overlooked because of their background, while recruiters will remain overwhelmed by large applicant pools. We often see talented people whose GitHub work reflects strong ability, but their resumes do not capture that skill. At the same time, recruiters will not have the time to manually examine every project or verify every claim.

This inspired our central question: What if AI could help recruiters evaluate talent fairly, based only on evidence and skill?

FairHire will be our answer to that question, designed to support unbiased hiring and align with the IndiaAI mission of responsible, equitable AI.

What it does

FairHire will function as an AI powered hiring assistant for recruiters. It will remove bias from resume screening, analyze GitHub projects, evaluate LinkedIn profiles and generate transparent, skills only candidate rankings.

It will include the following capabilities:

Skill extraction from resumes, LinkedIn and GitHub
Bias free resume screening
AI based GitHub project review and scoring
A unified candidate profile view
Interview analysis based on content
Fair shortlist generation and JD bias analysis

FairHire will help recruiters make fast, fair and evidence based decisions.

How we will build it

We plan to design FairHire as a modular, AI driven system with three main components.

Data Ingestion Layer This layer will use APIs to collect resumes, LinkedIn data, GitHub repositories and metadata. GitHub analysis will rely on static code analysis, repository level metrics and AI generated summaries.
AI Reasoning Layer This will include:

skill extraction
project evaluation
resume anonymization
bias detection
fairness constrained ranking

We intend to use LLM based models for text understanding and scoring, supported by deterministic fairness rules.

Recruiter Dashboard We will build a clean, actionable interface that displays

verified skills
GitHub project strength
ranking explanations
JD bias suggestions

Our goal is to make AI insights simple and trustworthy for recruiters.

Challenges we expect to face

We anticipate several challenges as we build FairHire.

Designing a fair ranking system that uses only skills and evidence without allowing hidden bias.
Ensuring the GitHub review engine evaluates complexity and code quality accurately.
Merging resume, LinkedIn and GitHub data into consistent candidate profiles.
Handling variations in commit patterns, documentation styles and coding practices.
Maintaining transparency and explainability across all AI decisions.

Accomplishments we expect to be proud of

We aim to achieve the following milestones.

A reliable GitHub auto review system that can assess project complexity, code quality and documentation depth.
A bias free screening workflow that operates without demographic influence.
Clear and intuitive visualizations such as the Skill Graph and GitHub Score Breakdown.
Strong alignment with IndiaAI principles of responsible and transparent AI.

What we expect to learn

We expect to learn how to build fairness into AI from the ground up. We expect to learn how to combine code analysis, text analysis and fairness constraints in a practical system. We expect to deepen our understanding of multi source data integration, explainable scoring methods and recruiter friendly UX design.

We also expect to learn how subtle design choices can influence fairness and recruiter trust.