IT 16 - AI-Driven Employee Performance and Pay Equity HR Platform MVP Development

In today’s evolving workplace, fostering diversity, equity, and inclusion (DEI) is no longer just a corporate ideal, it’s a strategic necessity. Companies that prioritize fairness and inclusion see measurable benefits in employee satisfaction, innovation, and retention. However, achieving these goals in HR practices remains challenging due to unconscious biases, fragmented systems, and lack of actionable data. To help solve these challenges, Exera Solutions Inc., in collaboration with Centennial College’s AI Capstone Program, is developing an AI-powered Software-as-a-Service (SaaS) platform. This Minimum Viable Product (MVP) is designed to help organizations optimize employee performance management, close pay equity gaps, and reduce bias in hiring workflows.

At the heart of the platform is the Hiring Module, a smart, AI-assisted system that empowers HR professionals to create and manage inclusive, professional job postings. This module includes three core functionalities: smart autofill, bias analysis, and privacy-preserving AI integration.

The smart autofill feature leverages advanced language models to help HR teams jump-start the job description writing process. When a job title is entered, the AI suggests content for key sections such as position summaries, required and preferred skills, and workplace responsibilities. These AI-generated suggestions follow inclusive language standards from the outset and can be easily customized by HR administrators, ensuring both consistency and flexibility across postings. This not only saves time but helps maintain high-quality, unbiased organization wide.

Following autofill, the job description can be submitted for AI-powered bias analysis. The platform checks for problematic language across several categories, including gender, age, racial, academic, and experience-based biases. Terms that may unintentionally exclude or discourage certain groups from applying are flagged, and neutral alternatives are proposed. The AI’s recommendations are presented with a clear, side-by-side comparison view that allows HR professionals to selectively accept or modify suggestions. Importantly, human oversight is always preserved, keeping the final decision in the hands of the people—not the machine.

Recognizing the sensitivity of corporate information, the Hiring Module also integrates robust data privacy measures. All job descriptions are partially anonymized before being sent to the AI model—company names, specific locations, and proprietary identifiers are removed or replaced with generic placeholders. This ensures that no private or sensitive corporate information is exposed during the AI processing stage. Additionally, the platform is designed in full compliance with Canadian data privacy regulations such as PIPEDA and aligns with AI governance principles outlined in the Directive on Automated Decision-Making and emerging AI regulations under Canada’s Artificial Intelligence and Data Act (AIDA). The anonymization process not only protects privacy but also allows for safe storage and retraining of models using curated, compliant datasets.

Technically, the Hiring Module is built using React.js for the frontend and Node.js/Express.js for backend services. MongoDB is used for data persistence, while the Gemini API provides real-time AI functionality. The system is containerized using Docker and Kubernetes, making it scalable and production ready. Logging and monitoring are implemented to ensure consistent AI performance and ethical usage over time. Future iterations of the module will incorporate feedback loops to further improve the AI’s accuracy by learning from accepted or rejected bias suggestions.

Together, the features in the Hiring Module eliminate manual bottlenecks, reduce unconscious bias, and give organizations the tools to make recruitment more transparent and equitable—without compromising data security or oversight.