Developed a Faithfulness Testing Framework for the JudgmentLabs/judgeval repository, enabling quantitative evaluation of response faithfulness across language model competitors. The work involved data engineering and analysis using Python, with enhancements to cstone_data.csv through the addition of an is_hallucination column to support structured faithfulness assessment. Integrated libraries such as Patronus, Ragas, and JudgmentClient within a new faithfulness_testing.py module to automate the evaluation process. This feature-oriented contribution established a foundation for systematic LLM evaluation, facilitating future comparisons and iterations. The approach emphasized robust testing and data-driven methodologies, focusing on extensibility and reproducibility in large language model benchmarking workflows.