James Ding
Data Scientist
TN Visa Eligible · No Sponsorship RequiredMDSA from the University of Calgary. Background in private aviation, logistics, and agriculture — I understand operational data from the ground up. I build end-to-end ML systems: from messy raw data to deployed models that drive real decisions. Currently shipping production forecasting at AirSprint and open to Data Science, AI Engineer, and ML Engineer roles.
Featured Project
AI Judge — SCOTUS Verdict Backtester
Can AI predict Supreme Court rulings? RAG pipeline over 200 landmark SCOTUS cases with temporal leakage prevention — Claude 3.5 Haiku on AWS Bedrock, FAISS vector search, MLflow experiment tracking, and a Streamlit dashboard. 45.7% backtested accuracy.
Work Highlight
Flight Hour Forecasting — AirSprint Private Aviation
Built and deployed a Prophet forecasting model on 12 years of private aviation data (78K+ flights). Achieved 6% weekly MAPE in live production, reducing positioning costs by $490K+. Architected automated ingestion pipelines (FL3XX, Salesforce, Foreflight → Lambda → S3 → Snowflake) and QuickSight dashboards for executive stakeholders.