CV

University of Southern California

Master of Science in Computer Science

• Anaylsis of Algorithms
• 3D Graphics and Rendering
• Multimedia Systems Design
• Autonomous Cyber-Physical Systems
• Information Retrieval and Web Search Engines
• Advanced Mobile Devices and Game Consoles

National Yang Ming Chiao Tung University

Bachelor of Science in Computer Science

• Parallel Programming
• Computer System Administration
• Introduction to Database Systems
• Computer Vision for UAV Autopilot
• Advanced Programming in the UNIX Environment
• Data Structures and Object-oriented Programming

Advantech

Software Development Intern

• Modernized a ResNet-based image classification model on Ampere GPUs in PyTorch using configurable loaders
• Deployed a MobileNetV3-based Faster R-CNN object detector in PyTorch to an internal development portal
• Implemented a Detectron2 RetinaNet model for solder-joint inspection by converting LabelMe annotations to COCO format, enabling deployment on Windows in production

MugArchive

• Built a React app to track Starbucks Discovery Series ornaments on an interactive US map
• Implemented state hover highlighting and rendered city and special-location markers from coordinate data
• Added a collection dashboard with total count, category breakdown, and quick status visibility
• Delivered grid and map views with real-time search and category filters for states, cities, and special locations
• Automated GitHub Pages deployments with GitHub Actions, rebuilding only upon relevant source changes

ParkMarks

• Built a React app to explore and track US national parks with an interactive map and clickable markers
• Implemented a park details panel with an image, established date, visited status, visit dates, and star ratings
• Developed search and status filters plus a progress dashboard for visited count and completion percentage
• Added shareable deep links via URL state for direct navigation and browser history support
• Shipped updates to GitHub Pages using an artifact-based GitHub Actions pipeline with concurrency control

DreamMesh: Blender Plugin for AI-Driven 3D Scene Generation

• Created a Blender plugin driven by generative AI to automate 3D scene creation from text prompts
• Implemented Stable Diffusion image generation and integrated Stable Fast 3D to convert 2D images into meshes
• Accelerated model inference with CUDA and optimized JSON-based scene parsing for asset placement
• Collaborated on a four-person international team, using Git to build a modular pipeline covering text-to-mesh, scene assembly, rigging, animation, and 3D scene configuration workflows for simulation-like environments

Wi-Fi Rate Control on Ryu and OVS

• Built a system-level Wi-Fi rate control service on Linux using Ryu and Open vSwitch over OpenFlow and TCP/IP
• Automated 802.11 telemetry collection (MAC, IP, RSSI, SNR) and injected metrics into OpenFlow metadata
• Parsed metrics in the Ryu controller to apply SNR-based rate and MCS selection, enforcing QoS and flow rules
• Improved throughput and latency while reducing controller overhead via batched rule updates, demonstrating practical experience in distributed control systems

Ransomware Propagation and Payload

• Automated SSH dictionary-based login attempts via shell scripts to simulate a ransomware attack
• Overrode the default "cat" command to enable worm-like propagation across devices
• Encrypted JPEG files via a deployed payload and triggered a custom ransom pop-up message
• Coordinated a two-person team using Git for version control and collaboration

Accelerating K-means Clustering with Parallel Implementation

• Architected three parallel K-means algorithms for distributed execution using OpenMP, MPI, and a hybrid model
• Wrote shell scripts to automate dataset generation, runtime logging, and parallel execution across configurations
• Benchmarked large-scale datasets of 20 million points with MPI, achieving up to 7.4x speedup
• Collaborated with a three-person team on a Git-based codebase to evaluate parallel efficiency and scalability while profiling performance bottlenecks across architectures

Analysis of EEG using Deep Learning

• Coordinated with a team to design an EEG experiment using E-Prime to collect signals during index finger lifts
• Preprocessed EEG data with EEGLAB, including filtering, artifact rejection, and channel segmentation
• Developed and trained an EEGNet classifier in PyTorch to predict motor intent prior to finger movement, achieving approximately 72.5% accuracy through cross-validation and confusion matrix analysis

When BERT meets Google Comments

• Built a BERT-based classifier with PyTorch to predict the sentiment and topic categories of Google Maps reviews
• Crafted modular scripts for web scraping with BeautifulSoup, model training, and model inference
• Achieved approximately 90% accuracy on a test set labeled with Google star ratings
• Integrated the model into a PyQt5 GUI for real-time, user-facing sentiment and category prediction
• Worked on a three-person team and coordinated via Git