Currently
Interesting in materials science and machine learning
Specialized in
Machine LearningDeep LearningScientific ComputingMaterials InformaticsData EngineeringCloud ComputingFull Stack Development
Professional Experience
Northeast Materials Database (NEMAD) Project
Jan. 2024 - Present
- Led development of a scalable database with 26,000+ entries using ETL pipelines and automated data collection
- Architected and deployed machine learning pipeline achieving 90% accuracy in multi-class classification
- Developed ensemble models for regression tasks achieving R² of 0.86 and reducing prediction error by 40%
- Implemented feature engineering techniques resulting in 62 high-value material discoveries
- Built and deployed production-ready REST API and web interface (www.nemad.org) with 150+ monthly active users
PythonVue.jsFastAPISQLAWSDocker
GPTArticleExtractor: Large-Scale Data Mining Project
Jan. 2023 - Jan. 2024
- Designed and implemented an NLP pipeline leveraging LLMs to extract structured data from 22,000+ scientific papers
- Integrated YOLOv8 model for automated segmentation of figures and text regions in research papers
- Achieved 83% accuracy in automated information extraction, reducing manual processing time by 90%
- Engineered vector embedding system for semantic search, improving information retrieval accuracy by 45%
- Developed custom data validation framework ensuring 95%+ data quality
- Created scalable ETL pipeline processing 10000+ papers per hour
- Built interactive dashboard for real-time data visualization and analysis
PythonLangChainOpenAI APIComputer VisionPrompt EngineeringVector DBPostgreSQL
Deep Learning for Vector Field Reconstruction
Jan. 2022 – Jan. 2023
- Developed custom U-Net architecture with ResNet backbone for complex image-to-vector field mapping
- Designed multi-scale CNN architecture incorporating skip connections and residual blocks
- Implemented coordinate transformation pipeline converting polar to Cartesian coordinates for 3D reconstruction
- Designed and optimized bilinear interpolation algorithms for accurate vector field projection
- Built data processing system handling 10TB+ of simulation data using distributed computing
- Achieved 92% accuracy in vector field prediction using custom loss functions
- Reduced computation time by 75% through GPU optimization and parallel processing
- Created automated testing framework for model validation
PyTorchCUDANumPyOpenCVSciPyLinuxCNNsResNet
Job Comparison Android Application
Jan. 2023 - May. 2023
- Architected and developed Android application for comparing job offers using Java and Android Studio
- Implemented systematic design approach using UML diagrams for initial architecture planning
- Designed and built intuitive user interface following Material Design principles
- Established Git workflow with feature branching strategy for team collaboration
- Created comprehensive testing suite including unit and integration tests
- Led weekly team meetings and coordinated development efforts through GitHub
- Managed feature implementation through agile development process
JavaAndroid StudioGitUMLMaterial DesignJUnit
Large-Scale Computational Analysis Platform
Nov. 2018 - Oct. 2020
- Built automated workflow for high-throughput computational chemistry calculations
- Developed Python scripts for data preprocessing and feature extraction
- Created interactive visualization dashboard for real-time analysis
- Implemented version control and documentation system for reproducible research
- Reduced analysis time by 80% through process automation
PythonLinuxGitJupyterMatplotlib
Brazilian E-Commerce Analysis Project
Aug. 2023 - Nov. 2023
- Conducted comprehensive analysis of Olist dataset focusing on customer satisfaction and purchasing patterns
- Developed predictive models using regression analysis and sentiment analysis techniques
- Created interactive dashboards and visualizations using Tableau, D3.js, and Matplotlib
- Performed geographic distribution analysis to identify regional sales patterns and opportunities
- Engineered recommendation system algorithm improving customer product discovery
- Implemented data processing pipeline handling 100,000+ transaction records
- Built interactive web-based visualization platform for real-time data exploration
Python (pandas)SQLiteTableauD3.jsJavaScriptMachine Learning
Work Experience
Research Assistant
Aug. 2019 - Present
University of New Hampshire
Teaching Assistant
Aug. 2018 – Aug. 2019
University of New Hampshire
Research Assistant
Oct. 2015 - Jan. 2017
Stony Brook University
Technical Skills
Machine Learning/AI:
TensorFlowPyTorchScikit-learnDeep LearningNeural NetworksGANsLangChainVector DatabasesOpenAIOllama
Data Processing/Analysis:
Python (Numpy, Pandas)SQLPySparkBig Data AnalyticsETL PipelinesDatabricksApache Airflow
Cloud Computing:
AWSGCPDockerKubernetesTerraform
Data Visualization:
D3.jsMatplotlibTableauPlotlyStreamlit
Development Tools:
GitLinuxUbuntuNixOSBashCI/CD
Network & Security:
SSHSSL/TLSNetwork Protocols
Programming Languages:
PythonSQLJavaScriptREST APIsFastAPI
Database Systems:
PostgreSQLMongoDBRedis
Computer Vision & Image Processing:
OpenCVImage reconstructionCoordinate transformationsBilinear interpolation3D visualization
Education
Georgia Institute of Technology, Online
Computer Science - Master's Degree
University of New Hampshire, Durham
Computational Chemistry - PhD's Degree
Stony Brook University, New York
Materials Science and Engineering - Master's Degree
Zhengzhou University, China
Materials Chemistry, College of Materials Sciences and Engineering - Bachelor's Degree
Publications
Large Language Model-Driven Database for Thermoelectric Materials
Itani, S., Zhang, Y., & Zang, J.
arXiv preprint arXiv:2501.00564 (2024)
Northeast Materials Database (NEMAD): Enabling Discovery of High Transition Temperature Magnetic Compounds
Itani, S., Zhang, Y., & Zang, J.
arXiv preprint arXiv:2409.15675 (2024)
GPTArticleExtractor: An automated workflow for magnetic material database construction
Zhang, Y., Itani, S., Khanal, K., Okyere, E., Smith, G., Takahashi, K., & Zang, J.
Journal of Magnetism and Magnetic Materials, 597, 172001 (2024)
Three-dimensional magnetization reconstruction from electron optical phase images with physical constraints
Lyu, B., Zhao, S., Zhang, Y., Wang, W., Zheng, F., Dunin-Borkowski, R. E., Zang, J., & Du, H.
Science China Physics, Mechanics & Astronomy, 67(11), 1-11 (2024)
MagNet: machine learning enhanced three-dimensional magnetic reconstruction
Lyu, B., Zhao, S., Zhang, Y., Wang, W., Du, H., & Zang, J.
arXiv preprint arXiv:2210.03066 (2022)
Conferences
2025 Joint MMM-Intermag Conference, New Orleans
January 13-17, 2025Poster presentation: "Comprehensive Database of Magnetic Materials Using AI-Driven Methodologies"
IEEE AtC-AtG Magnetics Conference 2024
October 2, 2024Oral presentation: "Comprehensive Database of Magnetic Materials Using AI-Driven Methodologies"
2022 Fall meeting of the New England sections (NES) of APS
October 14, 2022Poster presentation: "MagNet: machine learning enhanced three-dimensional magnetic reconstruction"