During January 2025, Fillzpp enhanced the GPU monitoring test infrastructure for the leptonai/gpud repository by refactoring end-to-end tests to use mock implementations for NVML, nvidia-smi, and lspci. This approach, leveraging Go and Bash scripting, enabled more reliable and deterministic test scenarios, directly addressing test flakiness and reducing CI noise. By updating the Codecov configuration to exclude the e2e directory, Fillzpp improved the accuracy of coverage reporting and provided clearer signals of test quality over time. The work demonstrated depth in system integration and mocking, resulting in a more robust and maintainable GPU monitoring test pipeline.