The Hong Kong University of Science and Technology
BEng in Artificial Intelligence Dept. of Computer Science and Engineering
Jinghe “Daniel” Wang
I am currently a undergraduate student majoring in Artificial Intelligence. I speak both English and Chinese (Mandarin). To know more about my life, please visit my personal website.
My current interest includes LLM agents and harness, practical applications of AI in real-world scenarios, and Apple platforms native application development. My work spans AI workflows, machine learning, and spatial application for Apple platforms.
Education
Experience
Software Engineer Intern
Beijing Zhongkexunbo Communication Technology Co., Ltd.
- Developed AI-driven workflows using Dify and large language models to analyze and process SMS content.
- Wrote Python scripts for data preprocessing, message parsing, and automated text classification.
- Designed prompts to extract structured information from unstructured SMS messages.
- Evaluated and optimized model outputs to improve accuracy and processing efficiency.
Projects
One Thing — Belonging Memories
iOS and visionOS application
SwiftSwiftUIARKitRealityKitAI text generation
- Developed an iOS application for capturing personal objects and recording the stories associated with them.
- Used ARKit and RealityKit to capture and render 3D objects, and implemented 3D visualization on Apple Vision Pro (visionOS).
- Built the interface in SwiftUI and core application logic in Swift.
- Integrated AI text generation to expand and organize user-written stories into structured narratives.
- Collaborated on the application workflow and UI/UX design; received Third Prize in the 2025 Mobile Application Innovation Competition (North China Region).
Artificial Intelligence Algorithms Implementation
Introduction to Artificial Intelligence coursework
PythonNumPyPyTorch
- Implemented core AI algorithms from scratch using NumPy.
- Built and tested Naive Bayes, Multilayer Perceptron (MLP), and Convolutional Neural Network (CNN) models in Python.
- Used vectorized operations and matrix-based computation for training, prediction, and clustering workflows.
- Evaluated behavior on supervised and unsupervised learning tasks through experimental analysis.
- Trained and fine-tuned large language models (LLMs) for classification and generation tasks.
Technical skills
- Development Languages
- Python, C++, Swift, Java
- Technologies
- SwiftUI, Dify, scikit-learn, LLM Skills
- Libraries
- PyTorch, pandas, NumPy, Matplotlib
- Tools
- Xcode, VS Code, Git, LaTeX, Pi Agent