Optimizing Candidate-Job Fit Using a Constraint-Aware Genetic Algorithm Framework

Organizations aiming to hire the most suitable candidate for a specific job position face the complex challenge of distinguishing the best option among a large pool of candidates while also adhering to multiple constraints such as cost, diversity, role-specific quota, and candidate quality. Driven to solve this challenging problem, this work presents a data-driven optimization framework that leverages a Genetic Algorithm (GA) to automate and enhance the candidate selection process. Our methodology evaluates each candidate based on a fit score, derived from weighted attributes such as skills, experience, and alignment with job requirements, while key constraints, including budget limits, a minimum percentage of female candidates, required role coverage (e.g., engineers), and minimum fit thresholds, are integrated directly into the fitness evaluation. Solutions that violate these constraints are ignored, ensuring that the model retains only feasible and high-quality selections. Collectively, the GA operates on binary selection vectors representing subsets of candidates and iteratively evolves toward an optimal solution using fitness-based selection, crossover, and mutation operators. To evaluate GA performance, the model was tested using synthetic Human Resource (HR) data and successfully identified a subset of candidates that maximized overall fit while satisfying all constraints, validating the effectiveness of the proposed approach.