Mathematical Framework for Custom Reward Functions in Job Application Evaluation using Reinforcement Learning

Most of the traditional Applicant Tracking Systems (ATS) depend on strict matching using keywords, where candidates that are highly qualified are many times disqualified because of minor semantic differences. In this article, the two-stage process of developing a more comprehensive resume assessment system based on small language model that is trained with fewer than 600M parameters is introduced and fine-tuned by using GRPO with a unique-designed reward function.The initial stage is (SFT) Supervised Fine Tuning, which are use to create a strong base model with the ability to perceive resumes beyond superficial overlap of keywords. This SFT model is further-optimized in the second step with Reinforced Learning (RL) via GRPO with the help of multi-component based rewarding, which will not be considered as a commission of tokens matching.In the initial RL experiments, we found a severe difficulty in the shape of reward hacking: overly aggressive penalty terms resulted in unstable training dynamics and prohibitively negative model behaviour. This was solved by trial and error refinement of the reward, and careful training hyperparameter tuning, which led to a stable and controlled process of gentle polishing.GRPO-refined model shows high real-life performance, as it shows accuracy of 91% on unseen data used for testing. It has a high recall of 0.85 on the SELECTED class with a perfect precision of 1.0, which highlights its high reliability to be used in identifying qualified applicants. These findings demonstrate that an appropriately structured two-step fine-tuning pipeline can effectively be used to transfer a small language model into humanlike candidate evaluation, surpassing shortcoming of both traditional ATS systems and unrefined uses of reinforcement learning.