YUJIANG LI

Master's student in Computer Science (AI/ML track) with a background in Economics, and hands-on experience in data preparation for speech recognition systems. Currently working in a university hourly position focused on speech data annotation for non-native English children, supporting downstream machine learning model training. Strong foundation in Python-based machine learning, neural networks, and statistical analysis, with a clear interest in entry-level Machine Learning/Data roles.

Data Annotation Assistant (Hourly Position)

Southern Methodist University | Dallas, Tx

2026 - Present

• Performed detailed manual annotation of speech audio samples to support speech recognition and assessment systems for children whose first language is not English.
• Labeled fine-grained temporal segments within audio recordings, including background noise, adult speech, child speech, filler sounds (e.g., humming), and task-relevant responses.
• Ensured high-quality, consistent annotations to improve the reliability of downstream supervised learning and model training.
• Collaborated with researchers to follow annotation guidelines and maintain dataset consistency for future machine learning experiments.

Feedforward Neural Networks for Structured Data Classification

Southern Methodist University | Machine Learning Lab | Fall 2025

• Designed and trained feedforward neural network models using a sequential modeling approach in Python.
• Implemented end-to-end training workflows including data preprocessing, train/validation splits, forward propagation, and loss-based optimization.
• Analyzed the impact of network depth, activation functions, and learning rates on model performance and convergence behavior.
• Evaluated models using quantitative metrics and learning curves to diagnose underfitting and overfitting.
• Documented experiments in a reproducible Jupyter Notebook with clearly structured code and analysis.

Sequential Neural Networks for Time-Series / Sequence Modeling

Southern Methodist University | Deep Learning Lab | Fall 2025

• Implemented and trained sequential neural network models to handle ordered and temporal data using Python.
• Built models using a step-by-step pipeline including data preparation, sequence batching, forward propagation, and loss computation.
• Experimented with different network configurations and hyperparameters to analyze their effects on convergence and model performance.
• Evaluated model behavior through training and validation loss curves, focusing on stability, overfitting, and learning dynamics.
• Documented experiments and results in a reproducible Jupyter Notebook with clear code structure and analysis.

The Evolution and Application of Natural Language Processing (ChatGPT Case Study)

Online Summer Workshop | June 2023

• Studied the progression of NLP methods from rule-based and statistical approaches to neural networks and pretrained language models.
• Analyzed applications of NLP in business domains such as intelligent search, machine translation, and automated content analysis.

Interoperability Between Apple Vision Pro and Conversational AI Systems

Online Summer Workshop | Summer 2023

• Discussed potential interaction points between spatial computing platforms and conversational AI for intelligent user support.
• Reviewed global trends and regulatory considerations in artificial intelligence development.