Data Scientist & Analyst | MS in Business Analytics @ UMN Carlson

I build applied analytics and machine learning solutions across forecasting, causal inference, large-scale data, and LLMs.

My background spans economics, risk management, and e-commerce operations, which shaped how I approach analytics: not as modeling for its own sake, but as a way to turn complex data into decisions people can act on.

Python PySpark Spark LLM

Built an LLM-assisted Spark workflow to detect data-quality anomalies, diagnose root causes, and generate validated PySpark fixes for analyst review.

Highlights: Large-scale data quality · Automated diagnosis · LLM-generated remediation · PySpark validation

R Difference-in-Differences Econometrics Causal Inference

Estimated the causal impact of sponsored advertising using a natural experiment and Difference-in-Differences, separating true incremental impact from observational correlation.

Highlights: Experiment design · Parallel trends · Treatment effect estimation · Business ROI

Python AutoGen ChromaDB FastAPI Azure

Built a multi-agent LLM system for natural-language queries over medical records, with role-based access enforced through metadata-filtered retrieval.

Highlights: Multi-agent architecture · RAG · Vector search · Access control · Cloud deployment

Python R SARIMA LSTM Time Series

Analyzed and forecasted long-term consumer search interest in Buy Now Pay Later using statistical and machine-learning time-series approaches.

Highlights: Time-series analysis · SARIMA · LSTM · Forecast evaluation

Languages
Python · SQL · R

Data & Distributed Computing
Spark · PySpark · Databricks · Hadoop · Hive

Machine Learning & Statistics
scikit-learn · LightGBM · XGBoost · Statistical Modeling · Time Series Forecasting

Analytics
Causal Inference · A/B Testing · Anomaly Detection · Clustering · PCA

GenAI & NLP
LLM Workflows · RAG · Multi-Agent Systems · ChromaDB

BI & Visualization
Tableau · Power BI

I'm particularly interested in problems at the intersection of:

• Data Analytics & Decision Science — turning complex data into actionable business insights
• Forecasting & Machine Learning — building predictive systems for real-world decision-making
• Causal Inference & Experimentation — understanding what actually drives outcomes
• LLM & Data Systems — using generative AI to improve analytical and data workflows

My path into analytics started with economics and risk management, where I became interested in how people and organizations make decisions under uncertainty.

After working in e-commerce operations and seeing how strongly pricing, assortment, and commercial decisions depended on data, I pursued an MS in Business Analytics at the Carlson School of Management, University of Minnesota.

Across projects in manufacturing, construction, transportation, healthcare, and data infrastructure, one principle has stuck with me:

Good analysis isn't about sophistication for its own sake — it's about producing something people can actually use to make better decisions.

Outside of analytics, you'll usually find me at a coffee shop with a book or out on a long walk.

LinkedIn: Yung Hsuan Hsieh
Email: hsieh203@umn.edu