▪ Developed a document auditing system capable of handling millions of documents, enabling users to browse, view, and download documents.▪ Developed a synchronization system to seamlessly sync changes from the database to the document auditing system.▪ Developed a data visualization system that enables users to intuitively interpret and visualize their data on the platform.

▪ Developed a system for the problem of detecting shoplifting from security cameras. The system runs real-time at 25FPS for 3 video streams on an Nvidia RTX 3060 GPU.▪ Developed a system for the problem of detecting illegal trash dumping from security cameras. The system runs real-time at 10FPS for 16 video streams on an Nvidia Tesla T4 GPU.▪ Implemented data pipeline for automating the processing of newly collected video data.

▪ Conducted researches and experiments on choosing the best loss weights in Self-supervised Depth Estimation models.▪ Redesigned and refactored Monocular Depth Estimation codebase for training, testing, and inferring Monocular Depth Estimation models.▪ Implemented and reproduced research papers in Motion Segmentation, Monocular Depth Estimation, Optical Flow Estimation, Camera Motion Estimation.▪ Reviewed data labeling for traffic sign detection.

▪ Studied courses: Linear Algebra, Probability and Statistics, Machine Learning, Deep Learning, Computer Vision, Natural Language Processing.▪ Built vehicle's distance estimator using a Semantic Segmentation model and Depth Estimation model.▪ Built image super-resolution model using GANs.

▪ Built AI core for Object Detection application in jewelry store.▪ Built AI core for OCR application in paper factory.