Cuda optimized c++ implementation of the North Star algorithm for gravitational wave localization

Localizing gravitational waves requires algorithms capable ofprocessing large scales of data efficiently. The NorthStar pipelineoriginally written in python provides a substructure for this taskhowever it faces some challenges in speed and performance. Thisproject attempts to rewrite the NorthStar pipeline using C++ as wellas optimizing it using the CUDA libraries to essentially accelerate theGPU. Initially google colab was used to optimize the code using theCUDA libraries, but hardware compatibility issues ended up restrictingthe performance. The project also faced challenges in regard toalgorithmic outputs that did not fully align with the original pythoncode. To overcome these challenges a laboratory computer was usedinstead and AI tools were implemented to debug code and improveKernel execution. The resulting C++ rewrite successfully compiledand executed however it was only partially optimized as outputinconsistencies remained. Nevertheless, this project provided avaluable understanding on the complexity of CUDA basedoptimization as well as the difficulties that arise when trying to ensureconsistency across different programming languages to lay out thegroundwork of gravitational wave localization.