Gary Kong's Projects
This repository implements a persona-tailored Retrieval-Augmented Generation (RAG) system designed to address the diverse Generative AI (GenAI) needs of research and marketing teams.
This project uses graph-based algorithms to optimize the delivery network for a startup focused on prepackaged meals. It also operationalizes this as part of a broader expansion strategy for the company.
This project investigates the voting difficulties experienced by Democratic and Republican voters in the 2020 election, utilizing data from the American National Election Studies (ANES). By analyzing this dataset, the project aims to understand the extent of voting challenges across political affiliations.
This project explores the use of motor actigraphy data collected from wearable devices to classify patients with depression. Motor actigraphy provides valuable information about individuals' physical activity and inactivity patterns, which can be indicative of their mental health status.
This project demonstrates a fully functional prediction API that leverages the DistilBERT model from Hugging Face for sentiment analysis. It encompasses an end-to-end machine learning pipeline, including model packaging, API creation with FastAPI, result caching with Redis, containerization with Docker, and deployment to Azure using Kubernetes.
🦜🔗 Build context-aware reasoning applications
This project focuses on analyzing key features that influence customer approval of motorcycles, leveraging data from Bikez.com. By exploring the impact of features such as engine responsiveness (torque) and fuel capacity, we aim to provide insights that can guide motorcycle manufacturers in product design to enhance customer satisfaction.
This project leverages AI to classify chest X-rays into Normal, COVID-19 pneumonia and non-COVID pneumonia categories using HOG, Radiomics, and ResNet
This project introduces an innovative approach using T5 models to detoxify online text. It focuses on enhancing the capabilities of T5 through bidirectional training, data augmentation and lexically constrained decoding.