Worked on the JetBrainsRuntime repository to enhance the loop optimization subsystem by refactoring core components in C++. Focused on codebase cleanup and API simplification, the work involved removing unused parameters and fields to improve readability and reduce the API surface, making future maintenance safer and more efficient. Introduced a targeted regression test to address a specific bail-out scenario in loop optimization, thereby strengthening reliability and preventing assertion failures. Leveraged skills in C++ development, code refactoring, and compiler optimization, with additional attention to robust testing practices. These contributions improved maintainability and stability, laying a foundation for safer future optimizations.