Implemented a vector database to store and search for vector embeddings in CUDA C++

Added shared GPU memory buffer functionality to reduce excessive host/device memory copies when transferring data between system processes using CUDA C++, and UNIX system calls

Profiled and optimized CUDA kernels for various neural network layers to properly utilize the GPU architecture.

Improved pre-processing and detection stages for a YOLOv3 object detection system using TensorFlow.

Debugged and developed backend modules performing various file I/O tasks

Organized regular meetings with members

Negotiated club funding with administration