• Developed a feature extractor neural network system using TensorRT, CUDA C++ to process streams of data in real-time

• 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

• Created a full YOLOv8 object detection system including data pre-processing with C and Python, training via Keras, and deployment using CUDA C++ and TensorRT on a Nvidia Orin AGX

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

• Maintained and added features to GTK-based GUI applications in C

• 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

• Planned trips and events

• Organized regular meetings with members

• Negotiated club funding with administration

Managed club funds and inventory