HANA KIM
Senior Data Scientist
Seattle, WA · h.kim@email.com · (555) 214-7788
Summary
Senior data scientist with 7+ years building ML models that drive measurable revenue and efficiency gains. Led a churn-prediction model that saved $2.3M annually and a demand-forecasting pipeline used across 15 product lines.
Experience
Senior Data Scientist · Amazon
2021 - Present
- Built a customer churn-prediction model (XGBoost) that identified at-risk accounts 6 weeks earlier, saving $2.3M in annual retention costs
- Led a team of 4 data scientists to deploy a real-time demand-forecasting pipeline processing 40M+ events/day, cutting inventory overstock by 18%
- Redesigned the A/B testing framework, reducing experiment analysis time from 3 days to 4 hours
- Mentor 6 junior scientists in a quarterly model review, cutting review turnaround to 24 hours
Data Scientist · Expedia Group
2018 - 2021
- Developed a hotel-ranking recommendation model that increased booking conversion by 11%
- Automated a fraud-detection pipeline using scikit-learn, reducing false positives by 27%
- Shipped an NLP review-summarization service in TensorFlow over 9M reviews in 40 markets
- Migrated 30 batch jobs to Spark, cutting nightly runtime from 7 hours to 90 minutes
Data Analyst · Zillow Group
2016 - 2018
- Built 20+ SQL pipelines feeding a Tableau listings dashboard used by 60 analysts weekly
- Cleaned and validated 12M rows of property records in Python, cutting data errors by 35%
- Ran cohort analyses on 4 marketing channels, reallocating $400K of annual spend to top channels
Education
M.S. Statistics — University of Washington · 2018
Skills
Python, R, SQL, TensorFlow, PyTorch, Scikit-learn, XGBoost, A/B Testing, Pandas, Spark, Tableau, NLP
Certifications
AWS Certified Machine Learning Engineer - Associate, Amazon Web Services
Google Cloud Certified - Professional Data Engineer, Google Cloud
Databricks Certified Machine Learning Professional, Databricks
Projects
Seattle Transit Delay Model - PyTorch model predicting bus delays from 3 years of GTFS data
pandas-profiling contributor - 14 merged PRs adding memory-efficient dtype inference
Bayesian study group - ran 20 modeling sessions for 45 UW statistics alumni
Languages
English (Native), Korean (Native), Japanese (Conversational)
Interests
Open-source ML tooling, Kaggle competitions, Trail running, Film photography




