Mingji Li

Jersey City,NJ

Summary

Master’s student in Computer Science at Stevens Institute of Technology with hands-on experience in machine learning, deep learning, and natural language processing. Built multiple end-to-end ML projects, including protein function prediction using ProtBERT and NLP-based recommendation systems.

Skilled in Python, PyTorch, and data analysis, with practical experience in model training, feature engineering, and performance evaluation. Strong ability to quickly learn new technologies and apply them to real-world problems.

Work History

Course Project – NLP for Protein Function Predict

Stevens Institute of Technology

Hoboken, NJ

04.2025 - 05.2025

- Built an end-to-end protein function prediction pipeline using NLP and deep learning techniques
- Leveraged ProtBERT to generate embeddings from raw amino acid sequences
- Designed rule-based labeling system from UniProt text data to construct training dataset
- Trained an MLP classifier with class imbalance handling for multi-class prediction
- Evaluated model performance using accuracy and F1-score on real-world dataset (~17k samples)
- Developed a GUI application for real-time protein function prediction

Education

Master of Science - Computer Science

Stevens Institute of Technology

Hoboken, NJ

05-2026

Skills

Programming: Python, Java, C, C#

Machine Learning: Scikit-learn, Model Training, Classification

Deep Learning: PyTorch, Neural Networks

Natural Language Processing (NLP): ProtBERT, Text Processing

Data Analysis: Pandas, NumPy

Data Visualization: Matplotlib

Tools: Git, Jupyter Notebook

Concepts: Data Preprocessing, Feature Engineering, Model Evaluation (F1-score, Accuracy)

Languages

English

Professional Working

Chinese (Mandarin)

Native or Bilingual

Timeline

Course Project – NLP for Protein Function Predict

Stevens Institute of Technology

04.2025 - 05.2025

Master of Science - Computer Science

Stevens Institute of Technology