Son Le's Data Science Portfolio

This repository showcases hands-on data science and analytics projects across machine learning, experimentation, forecasting, business intelligence, and applied Python.

Why this portfolio stands out

• End-to-end project work: data extraction, cleaning, modeling, evaluation, and communication
• Business-focused problem solving across product, finance, and operations use cases
• Experience with both notebooks and deployable Python applications
• Evidence of practical delivery: dashboards, APIs, forecasting workflows, and Kaggle results

Featured projects

Project	Focus	Tools	Outcome
Data Engineering Journey	Data engineering portfolio spanning ETL, streaming, analytics engineering, and cloud workflows	Python, SQL, Bash, Kafka, Spark, PostgreSQL, MongoDB, Airflow, dbt	Demonstrates progression from core data skills to end-to-end pipeline design across multiple production-style projects
IBM RAG and Agentic AI Capstone Project	Retrieval-augmented generation and agentic AI workflows in a production-oriented capstone	Python, LLMs, RAG pipeline design, agent frameworks	Delivers a modern AI portfolio project focused on enterprise-style knowledge retrieval and agent orchestration
Stock Volatility Forecasting In Vietnam	ETL, time-series forecasting, and API development	Python, Pandas, SQLite, GARCH, FastAPI	Built a workflow that collects market data, trains a forecasting model, and serves predictions
Digital Rights Management	BI reporting and usage tracking	SQL, MySQL, Power BI / BI dashboarding	Built reporting to monitor DRM key usage and support forecasting and budgeting

Project categories

Machine learning

• Classification and predictive modeling
• Feature engineering and model comparison
• Model interpretation and decision support

Analytics and experimentation

• Exploratory data analysis
• Diagnostic analysis
• A/B testing and hypothesis testing

Forecasting and data products

• Time-series modeling
• ETL pipelines
• API-oriented analytics workflows

Business intelligence

• SQL-based data extraction
• Dashboard design
• Operational reporting

Repository structure

• Data Engineering Journey/ — featured overview of a separate repository focused on ETL, streaming, analytics engineering, and cloud data pipelines
• IBM RAG and Agentic AI Capstone Project/ — featured overview of an IBM capstone project repository focused on RAG and agentic AI system design
• Stock Volatility Forecasting In Vietnam/ — volatility forecasting workflow and supporting Python modules
• AB Testing Advertising Campaign/ — experiment analysis notebook and documentation
• Digital Rights Management/ — BI dashboard assets and supporting project files
• Titanic Survival Prediction/, Fraud Detection/, Recipe_Site_Traffic_Prediction/, Shop_the_Look_Recommender/ and more — additional foundational and exploratory portfolio work

Core tools used across projects

Python Pandas scikit-learn SQL SQLite MySQL PostgreSQL FastAPI Kafka Spark dbt Airflow MongoDB Power BI Data Visualization Hypothesis Testing Time-Series Forecasting

What recruiters and hiring managers can expect

• A portfolio built around real analytical workflows, not just isolated models
• Projects that connect technical execution with measurable business questions
• Clear evidence of curiosity across multiple data domains and problem types

Contact

If you are hiring for data science, machine learning, analytics, or BI roles, feel free to reach out:

• Email: haison19952013@gmail.com
• GitHub: haison19952013

License

This repository is licensed under the GNU General Public License v3.0.