📫 Contact: ptand@uic.edu | LinkedIn
I am a Master of Science in Computer Science student at the University of Illinois Chicago, passionate about software development, data science, AI, and cloud technologies. I specialize in building data-driven solutions and experimenting with GenAI tools.
Description:
Preprocessed over 26,000 rows and 305 columns of Zillow housing data and mortgage rates spanning two decades. Developed predictive models to estimate property appreciation across various U.S. states.
Skills & Tools Used:
- Python
- Pandas
- NumPy
- ARIMA
- Plotly
Outcome:
Built ARIMA models for 50 U.S. states to predict property value trends for timeframes of 1 month, 1 quarter, and 1 year, achieving an accuracy of 68%. Interactive visualizations were created to showcase trends, which can be accessed through the report.
Project Report:
Download the Property Estimation Project Report
Acknowledgements:
- Zillow Housing Data
- Various open-source libraries like Pandas, NumPy, and ARIMA
2. 🔗 DualDB Connect API
Description:
Built a RESTful API using Flask to connect MongoDB and MySQL, efficiently managing 100,000+ records. Optimized multiple CRUD endpoints to handle high request loads.
Skills & Tools Used:
- Flask
- MongoDB
- MySQL
- Python
Outcome:
Improved API response times by 30% and supported 500+ requests/second.
Description:
Developed a desktop application using PyQt to facilitate team collaboration, including features for video streaming, screen sharing, to-do lists, and chat groups.
Skills & Tools Used:
- PyQt
- SQLite
- Python
Outcome:
Enhanced team productivity by providing personalized scheduling suggestions and improving communication.
Description:
Deployed a scalable cloud application using AWS services such as CodePipeline, Elastic Beanstalk, and CloudFormation. Implemented auto-scaling, load balancing, and data security measures.
Skills & Tools Used:
- AWS
- Docker
- ECS
- S3
- Auto-Scaling Groups
- IAM
Outcome:
Ensured high availability, scalable performance, and robust security for sensitive financial data.
Description:
Built an ETL pipeline to forecast hotel cancellations using logistic regression and decision trees. Created a Flask-based dashboard to visualize cancellation trends.
Skills & Tools Used:
- Python
- PostgreSQL
- Flask
- Seaborn
Outcome:
Reduced data processing time by 25% and provided actionable insights for hotel management.
- Programming: Python, Flask
- Data Visualization: Matplotlib, Seaborn, Plotly, Power BI
- Cloud Platforms: AWS (CodePipeline, Elastic Beanstalk, S3, ECS)
- Databases: MongoDB, MySQL, PostgreSQL
- AI/ML: ARIMA, Logistic Regression, Decision Trees
- Tools: Docker, PyQt, Figma, Microsoft Office