Developed a Bias and Fairness Metrics Test Suite for the IFRI-AI-Classes/ifri_mini_ml_lib repository, focusing on systematic evaluation of machine learning model bias. The suite implemented tests for selection rate, demographic parity, and equalized odds, addressing both basic scenarios and complex edge cases such as empty predictions and varying positive label values. Emphasizing robust model governance, the work enhanced testing reliability by including weighted sample handling. The project was delivered in Python, leveraging skills in fairness in AI, machine learning, and testing. During this period, the primary focus remained on feature development and comprehensive test coverage rather than bug fixes.