I am a Data Scientist based in Cary, NC. I specialize in bridging the gap between advanced statistical modeling, causal inference, and machine learning to solve complex problems in both Marketing Science and Biomedical Research.

• 🔭 Currently: Consultant, building models to quantify incremental impacts and optimize budget decisions.
• 🎓 Education: Ph.D. in Biochemistry(structure biology) (HKUST), M.S. in Statistics (U of Arizona), and M.S. in Computer Science (Georgia Tech).
• 🔬 Research: Author of 30+ peer-reviewed publications with 2100+ citations.

Machine Learning & AI

• Deep Learning: PyTorch, CNN, Transformers, Seq2Seq
• LLMs: Fine-tuning (LoRA), Al Agents, Function Calling, LangChain
• Domains: Image Processing, Natural Language Processing

Causal Inference & Statistics

• Methods: Difference-in-Differences (DiD), Synthetic Controls, TBR-MM, A/B Testing, Matching Markets,Marketing Mix Modeling
• Modeling: Bayesian Modeling (MCMC), Time-Series Analysis, Survival Analysis, Mixed Models

Data Engineering & Tools

• Languages: Python, R, SQL, SAS
• Big Data: PySpark/Hive, Google Cloud Platform (GCP)
• Workflows: Nextflow, ETL Pipelines, Flask, R Shiny

I currently work on optimizing geo-pairs and driving multi-million-dollar budget decisions using causal analysis. My work involves:

• Implementing clustering algorithms to identify optimal geo-pair selections.
• Applying Time-Based Regression Matched Markets (TBR-MM) models.
• Utilizing Bayesian Inference for robust uncertainty quantification.

With over 8 years of experience in high-dimensional omics, I have:

• Led analytics collaborations for University of Arizona, processing 1,000+ animal/human datasets.
• Applied ML to real-world evidence datasets (NACC, ADNI) to uncover drug repositioning opportunities.
• Developed computational pattern-matching methods for target identification at HKUST.

• 🌐 Personal Blog: raymondshang.github.io
• 💼 LinkedIn: linkedin.com/in/yuanshang2020
• 📜 Google Scholar: Yuan Shang Publications
• 📧 Email: Shangyuan5000@gmail.com